In fact, according to analyst firms, measuring and reporting AI value was a top barrier to implementing AI initiatives. Some CDOs allow experimentation without a full business case; however, putting an AI model into production most often requires a clear picture of the return. 

The struggle to measure business value has been in the headlines lately. Investors worry about an AI bubble: Can growth among AI vendors continue, and can customers sustain their rates of investment? “[W]e question how it is going to work and whether it will generate sufficient return on capital,” worries one industry analyst. That remains the question du jour: “Is all of this actually worth anything?” We know it is. Many of our Snowflake customers are proving it.

Bottom line: It will be up to data and AI leaders to deliver and measure the business impact of their AI initiatives and the underlying data. Many already are. 

The effort and impact of these leaders shouldn’t go unnoticed or unappreciated, so Snowflake is thrilled to announce our Value-Focused Data and AI Leaders to Watch list. To help spotlight the contributions of these leaders, Snowflake has identified 50 individuals who deliver value to their organizations and have established processes to monitor and measure the impact on the business systematically. 

While it’s not a comprehensive list of the CDOs Snowflake believes are driving change within their organizations, we want to highlight a few who embody the qualities we consider critical to success. These leaders are progressive thinkers who have demonstrated the agility required to adapt to the fast-moving innovation around data and AI and recognize the imperative of delivering business value. 

The following individuals have established mechanisms to measure and monitor the business value derived from innovation and investments in data and AI. 

AI is creating exciting opportunities today for sports and betting organizations looking for ways to beat the competition by enhancing their personalized fan engagement strategies, creating new monetization opportunities, and boosting existing league and team operations strategies using the best tools available.

As sports organizations strive to meet rising fan expectations and stay competitive, leaders across the industry are leveraging AI to win.

Know your opponent

Establishing winning strategies on and off the field requires a team effort. It also takes research, knowledge, strategy and, most of all, time.

Sports organizations deploy significant resources to collect mountains of data on fans, players and more. Legacy systems, old approaches and segmented data can make it challenging to mine and maximize results from structured data, like ticket or merchandise purchase transactions, and unstructured data, like game footage.

Employing the latest in AI and machine learning (ML) delivers a deep bench of virtual teammates that can deploy predictive analytics to accelerate, streamline and improve the quality of the data, or allow sports organizations to leverage technology and information to give them something everyone in both the front and back offices craves: a winning edge.

Beef up your playbook

Think of a modern data strategy as your game plan: It’s enabling organizations to ask honest questions about the state of their data, strengthen their weaknesses, elevate their strengths and — crucially — realize cost savings and drive efficiencies.

Once teams get their data house in order, gen AI is being used to open up new opportunities to create more meaningful experiences for fans, drive more value for sponsors, and improve player and team performance. The growing gen AI market offers tremendous opportunities across the sports ecosystem with a modern data strategy the top sports organizations are already deploying:

• Personalized fan engagement and marketing: Enhanced loyalty programs, customized content recommendations and tailored marketing campaigns standing on the shoulders of ML, gen AI and Fan 360 are allowing sports organizations to find — and thrill — wider audiences through personalized experiences.

• Automated content creation and enhancement: Sports organizations are using gen AI to uplevel broadcast and streaming experiences for fans. They are automating alternate commentary options and the translation of broadcasts into multiple languages.  They are also automating data-driven visualizations that are incorporated into broadcasts to engage fans from the most casual to the most passionate.

• Player and team performance analysis: Leading organizations are maximizing performance and protecting the well-being of athletes by using gen AI and ML to analyze player data, including physical metrics and past performances, to understand individual player strengths and weaknesses and develop personalized training regimens. They’re also leveraging automated insights toward roster-building strategies and using gen AI to analyze opposing teams and simulate games as part of developing game plans.  

• Sponsorship activation: Optimizing sponsorship implementation is becoming more important as increasing sports media rights fees are causing sponsorships to get more expensive. Gen AI, along with ML, is enabling thoughtful and rewarding implementations to help organizations ensure sponsors remain top-of-mind for audiences.

• Key metrics for sponsorship measurement: Sponsors are seeing the bang for their buck, using advanced analytics to track impressions, measure reach and brand awareness, analyze demographics, assess brand sentiment, offer competitor analysis and sales metrics, and more.

Changing the game with a modern data and AI strategy

In this dynamic landscape, sports organizations harnessing the power of traditional AI/ML, gen AI and data, and adopting a modern data strategy are the ones with the winning team. 

In any sport, sometimes a team is its own toughest opponent. Both in the front office and on the field, that can look like:

• Legacy technology platforms or tools that are no longer agile

• A shallow bench of staff tasked with managing a massive data lift

• A lack of hardware and software tools to help the team be flexible and nimble

Organizations updating their playbook to include a modern data strategy enables them to create a connected data ecosystem that gives their team an edge over the competition.

The leading modern data cloud platforms lend crucial strategy — and strengths — to your organization:

• Reduced complexity: Sports organizations, like all other organizations, often grapple with managing complex, customized AI solutions. Bridging the gap between technical data teams and business and sports operations teams is a challenge. A data cloud’s ability to leverage AI to evaluate and interpret sprawling yet specific data is empowering sports organizations to fully utilize AI-driven tools for performance analysis or fan engagement. 

• Lower total cost of ownership: Handling vast amounts of data from player and fan interactions for AI models can be expensive and time-consuming. Modern data cloud platforms, which allow you to bring all your data into one secure, governed place, are set up so you only pay for what you use. And they’re lower risk because they manage the infrastructure risk for you. 

• Easy to use: Modern data clouds are fully managed platforms, removing the overhead of managing infrastructure. They’re flexible, allowing for agility and efficiency. And they offer faster implementation — and success. 

• Baked-in security and governance: Sports organizations handle sensitive data — including fan data and proprietary performance strategies — and that must be privacy-protected. Ensuring the platform has proper oversight of AI tools used to analyze player stats and engage fans is essential, as is adhering to regulations related to data privacy and AI ethics.

Are you ready to play?

These are just a few of the ways gen AI and a robust data strategy are helping sports organizations grow and mobilize their fans, increase revenue, and improve on-the-field operations. Teams investing in a modern data strategy are capitalizing on new opportunities, making their data more complete and actionable, and staying ahead of the competition.

Ready to learn more about tackling the challenge? Download our ebook, Game Changer: How Gen AI is Revolutionalizing Sports.

At Snowflake Summit this summer, an executive from a major manufacturing company reflected wistfully, “If we only knew what we knew.” In other words, imagine all the things we could do if we could derive insights from all the data we have collected, all that we know about our customers or our products. 

However, the challenge is knowing where to start. And, the answer lies in the prework: It’s what you do before you even fire up the tool. Imagine building a house without the plans. But even before the architect puts pencil to paper, there is a wish list of features for the dream house. How do you envision your life in the house? “Well, we’d like to take advantage of the gorgeous view the property has. To maximize the view, we’d like large picture windows or a wraparound porch. We’d also like to ensure the livability of the house with plenty of storage, so don’t forget the closet space.” Having recently gone through this process, I know it well. My wish list was long. 

Then the hard part comes. You’ve got a budget and likely a set of rules to follow, such as the building code or the neighborhood ordinances. Constraints require making hard choices and prioritizing certain features over others. Ideally, you can start with a solid foundation and key features and allow for some of the others to be added in later. And maybe you’ll have to learn how to do some things yourself so that your desired features become DIY projects. But likely some of these projects will not see the light of day.

Sound familiar? Maybe you’ve built a house. Or maybe you’ve built a data and AI strategy and started to use it. You’ve evangelized the opportunities and experimented with a number of pilots, but now it’s time to determine what makes the cut.

The pivot point from plan to production: Prioritization

A recent Harvard Business Review article reports that 80% of AI projects don’t make it into production. That figure has been used to illustrate failure. However, a closer look reveals that some of those 80% were sidelined intentionally. The first step in moving from experimentation to implementation is selecting which projects or products should move forward. As one CDO described it to me years ago, the goal of prioritization is to ensure the view is worth the climb.

Ideas come from all parts of the business, and that’s a good thing. Diversity of ideas should be encouraged. AI sandboxes and hackathons encourage experimentation. Ultimately, however, those ideas must be put to the test. A rigorous and transparent prioritization framework helps ensure that proposed projects and products align with business goals and that it’s realistic for the team to create them. 

The view. In the matrix above, the y-axis shows alignment with strategic business goals, broadly speaking. This is where you’d estimate the potential “view.” 

• Is the initiative aligned to business goals? Priority must be given to projects or products that have a direct link to specific business goals. Is there a business sponsor for the initiative? Data products won’t deliver value unless they drive action — that is, unless end users adopt them. This requires collaboration between business and data teams, and education on both sides about what is required and what is possible. 

• Can multiple business units benefit? Many data teams emphasize reuse as a requirement for prioritization: Value increases as more business units use the AI model or data product. For example, imagine that an electronics manufacturer wants to understand how its devices are used and by whom. A "product usage” data product could provide a customer 360 to see which products a customer was using and a product 360 to see which customers were using a particular product. 

• What is the anticipated return? This starts with identifying the metrics to be measured, ideally in terms of business value, and putting an estimated stake in the ground. Prioritization requires comparison across competing initiatives in concrete terms. 

The climb. The x-axis reflects the complexity and feasibility of a specific initiative to determine whether it is achievable in terms of resources and risks. This is where you’d estimate the required “climb.”

• Is the data available? The most critical requirement is the data to train the model. Is internal data easily available and accessible? Does the model require transformation or access to unstructured data? Would adequate training require external data, such as partner data or other third-party data, to mitigate risks of bias or hallucination? 

• What skills and tools are required? Here is where you have to be realistic about whether you have the resources to deliver, and what it takes to get there. An AI initiative shouldn’t feel like a sci-fi movie. 

• Are there risks involved or other concerns? For starters, the EU AI Act takes a risk-based approach to regulation and classifies AI systems into four different risk levels: unacceptable, high, limited and minimal risk. A number of tools can help assess risk level and provide guidance on the relevant regulations and requirements. 

• What is the cost to build and deploy? Ultimately, this is a business case and requires an estimation of the costs to build, deploy and maintain the AI model over time. These estimates should also include technology and data acquisition as well as training required. It’s not just about building an application. 

The end result is a matrix in which each potential product or project can be placed and evaluated in terms of the predicted view and the required climb.

The two-by-two matrix yields the following categories: 

• High-value quick wins: Great view, short and easy climb. These are the initiatives that fully align with strategic business objectives and are considered either less complex or highly feasible. 

• Don’t bother: Limited view, tough slog. These are considered harder and lower-potential-value initiatives that are not necessarily worth the effort. 

• Long-term investments: Great view, but long and steep climb.These initiatives are expected to deliver significant value, but they are considered more complex or come with significantly more risk or resource requirements. These might be broken down into smaller initiatives or component data products that could later be aggregated to deliver the full value. 

• Investigate further: Limited view, but relatively short and easy climb. These initiatives are expected to deliver some value (or they wouldn’t be proposed) but are less strategically aligned. However, they are considered relatively easy to deliver. It’s more of a “could do” but only with remaining time or resources, hence the lower priority.

This isn’t an optional process. The move from proof of concept to product/project involves hard choices. A formal prioritization framework ensures that initiatives are evaluated equally and transparently, that initiatives support business strategy, that initiatives are feasible for the organization and that resource requirements and expected outcomes align.

To hear more about how Snowflake customers have traveled the AI journey from evangelism and experimentation to operationalization and transformation, check out “The Data Executive’s Guide to Effective AI.”

Gain direct access to data for AI app development and analytics with Microsoft Power Platform, Microsoft Dynamics 365 and Snowflake  

We jointly announced an expanded partnership to enable bidirectional access between Microsoft Dynamics 365, Microsoft Power Platform and Snowflake. This access enables developers and business users to leverage data in Snowflake directly from the Power Platform and Dynamics 365. Developers can now build business applications with their data in Snowflake, removing the operational burden of managing custom workflows and shortening time to value. Coming soon, users will also be able to access their data from Dynamics 365 and Power Platform in Snowflake to take full advantage of the AI Data Cloud to improve analytic insights and harness the power of enterprise AI. 

The connector between Dataverse — the data storage and management layer of Microsoft Power Platform — and Snowflake will let users unite data from Dynamics 365, Microsoft Power Platform and Snowflake, simplifying data collaboration, enhancing business insights and harnessing the power of AI for their business needs. 

See how it works and hear from SKF, a global manufacturing leader, on how they’ve implemented the connector and are experiencing faster time-to-value by removing some of the operational complexities of data integration.

Revolutionize operations and enhance customer experiences with Snowflake Cortex AI and Microsoft Fabric

Marrying structured and unstructured data in the same environment can be seen as a challenge without the right resources and technical skill set. That’s why we developed Snowflake Cortex Search and Cortex Analyst (public preview). Now, all technical and nontechnical users can use valuable data that was previously too complex to work with to easily make data-driven decisions much more efficiently. 

At Ignite, we showed an example of how Snowflake Cortex AI works with Microsoft Fabric for a seamless workflow, starting with data ingestion in Snowflake and ending with insights presented in Microsoft’s easy-to-use Fabric platform. We demoed how a healthcare provider can use Snowflake Cortex AI to easily and securely uncover critical insights from vast amounts of patient data in documents. Seamlessly integrated with Microsoft Fabric, this powerful solution delivered fast results while maintaining top-tier security and compliance. 

Join our virtual hands-on lab to see how this works and try it yourself.

Interoperate with Fabric OneLake through Apache Iceberg

This past spring, we announced an expanded partnership with Microsoft, enabling customers to interoperate between Microsoft Fabric and Snowflake through Apache Iceberg to flexibly and easily connect and work across both platforms. Since then, we’ve made great progress to put this integration into action. Now in public preview, we support writing to Microsoft OneLake as the storage location for your data lakehouse to easily work with one copy of data from both platforms, reducing storage and pipeline costs.

The capability to write Iceberg Tables in OneLake and read them through Fabric services is just the start. Soon, Snowflake will be able to read any Fabric data artifact in OneLake, stored physically or virtually through shortcuts, and Snowflake’s engine will operate on data in OneLake. We’re working diligently to bring these capabilities to life to deliver a fully seamless, bidirectional data access experience.

Watch this demo and use this step-by-step quickstart to see how it works. 

Seeing it all in action with Power BI 

We had a lot of great conversations with many customers who are using Microsoft Power BI with Snowflake so that their organizations can visualize business impact and improve analytics and knowledge sharing across their organizations. We discussed how to use Direct Query Mode in Power BI to optimize analytics for sales teams, improve operations and enhance customer experiences. 

Try it out with this new quickstart for end-to-end analytics. 

For more resources, head to our Microsoft Partnership page.

This approach is all about helping you get the best price for performance with Snowflake. Plus, with our consumption-based pricing model, these performance boosts can lead to real cost savings for you.

Snowflake Performance Index results

Our dedication to your success drives us to continually measure and enhance the performance you experience with Snowflake. The Snowflake Performance Index (SPI) tracks these real-world improvements over time. Instead of using synthetic benchmarks for performance comparisons, we measure our enhancements using real customer data on production workloads. 

This means the SPI reflects genuine improvements that make a difference in your day-to-day operations. Since we launched the SPI in August 2022, the average query duration for stable workloads has now improved by 40%. In the last 12 months alone, the SPI has seen a 20% improvement.

Latest performance improvements tracked by the SPI

Over the past 12 months, we've introduced several significant improvements — mostly happening automatically, without needing any configuration or additional effort to modify code.

Query execution improvement: We’ve invested effort to continue reducing execution times and handling complex query patterns more effectively. Examples include optimizing join queries, automatically handling skew and expanding support for Top-K pruning to improve performance for queries with specific aggregation and filtering patterns. These updates help your queries run faster, even as workloads grow in complexity.

Data ingestion and replication: We reduced the time spent on metadata replication, we made cloning faster, and we optimized the ingestion of large data sets to help you bring data into Snowflake faster and more reliably, streamlining your workflows and pipelines.

Adaptive optimization: We launched a number of adaptive optimizations to make Snowflake smarter at choosing the best strategies for query execution. For instance, we've expanded Top-K pruning to include a broader range of queries and refined the optimizer's ability to make intelligent join-order decisions, benefitting you from faster, more efficient query planning.

Platform efficiency: We continued to enhance the platform's overall reliability and speed. For example, we reduced the time required for cloning operations and made compression more efficient, reducing resource consumption and enabling smoother system operations.

These are just a few examples of how we're making Snowflake faster and more efficient for you. We're committed to keeping this momentum going. By continuously investing in performance enhancements, we're dedicated to helping you get more value from Snowflake while reducing your operational costs over time.

To learn more about the latest improvements, check out our performance release notes and visit the SPI website.

*Based on internal Snowflake data, average query duration for customers’ stable workloads improved by 40% from August 25, 2022, to October 31, 2024. To calculate SPI, we identify a group of customer workloads that are stable and comparable in both amount of queries and data processed over the period presented. Reduction in query duration resulted from a combination of factors, including hardware and software improvements and customer optimizations. Improvement in query duration metrics are rounded to the nearest hundredth.

With Snowflake’s easy-to-use, unified AI and data platform, businesses are removing the manual drudgery, bottlenecks and error-prone labor that stymie productivity, and they are using generative AI to deliver new insights — and revenue streams. But what does that look like in practice?  

We’ve gathered some innovative generative AI solutions that our customers are using in production today. Their stories demonstrate how Snowflake and Cortex AI are putting gen AI goals within reach and driving business value along the way.

Johnnie-O improves accuracy of geocoding address data to better serve customers

Like many largely ecommerce businesses, the East-Coast-prep-meets-West-Coast-casual clothing brand Johnnie-O understands the value in a simple shipping address. Just a few lines of text can provide powerful demographic insights into the company’s customers when linked to data from the U.S. Census Bureau — information like average household income in the area, percentage of people with degrees, employment rates, races and ethnicities and so on. By using this data not only directly from website orders but from wholesalers and dropshippers, Johnnie-O can begin to understand its customer base better and consequently target its marketing efforts more effectively. 

But the company had one problem: A significant number of collected addresses could not be geocoded, preventing the team from accessing relevant customer data. Typically, the company runs raw address data through an application that delivers geographic coordinates, which then makes it easy to link to census data. But for Johnnie-O, many of these addresses failed for a variety of reasons, which could be as small as a typo or information in the wrong field. So instead of manually cleaning up these hundreds of thousands of data points, the company looked to Cortex AI to automatically reformat the messy address data. After feeding these incorrect addresses into Cortex AI using a Llama LLM, Johnnie-O immediately slashed its failure rate to just 2%. 

Now the company can run its market segmentation algorithms with confidence, knowing there are no significant holes in the data powering them. And to make this feat even more impressive, it was essentially built by just one person: Johnnie-O’s Analytics Engineer, Ricardo Lopez. “Cortex AI is so easy to use and implement, especially because all of our data is already in Snowflake,” he says. “Snowflake and Cortex AI have become the center of everything for us.”

Using Cortex AI, healthcare staffing agency IntelyCare is confident that job postings are no longer falling through the cracks 

Staffing jobs in healthcare is critical to a well-functioning medical system; it is also becoming increasingly complex, as many states anticipate facing nursing shortages in coming years. IntelyCare provides a comprehensive platform that helps match healthcare organizations with qualified nursing professionals for open positions, whether they be permanent roles, travel assignments or per diem shifts. With hundreds of thousands of open positions nationwide at any given time, the task of filling those roles begins with organizing job postings. 

While IntelyCare has direct relationships with many organizations, many of the largest healthcare systems prefer to post openings through a vendor management system (VMS), which can be accessed only by vetted agencies like IntelyCare. To be able to include these opportunities in its database and app, however, IntelyCare must process each post in a way that is organized and relatively uniform. That, of course, presents a challenge, given that each VMS adheres to its own system of standard practices. It’s not uncommon for fields to be left blank on some posts or for bodies of text to be unintelligible to IntelyCare’s internal tools; as a result, more than 30% of job posts were getting lost in processing. 

So IntelyCare began using LLMs in Cortex AI to quickly extract pertinent information, both simple and complex, from these thousands of posts: specialty, pay range, specialized years of experience required, whether local applicants can apply for travel positions and so on. Then IntelyCare can organize the posts in thoughtful ways and without fear of losing opportunities because of, say, incompatible formatting. “We’ve basically pushed that 30% of lost job posts down to zero,” says IntelyCare’s VP of Data Science, Benjamin Tengelsen. Not only has that enhanced the user experience for applicants, but it’s also lightened the burden on IntelyCare’s recruiters, who would tirelessly sift through individual postings to find the best matches for their candidates. 

Similarly, Cortex AI also helps Intelycare manage and process the constant stream of postings it fetches from public job boards. The team, for instance, had built elaborate pipelines to appropriately add tags to jobs for easy categorization; adding a new tag would require building and training new models, complex orchestration and frequent maintenance. “We can now replace thousands of lines of mangled Python code with a single Cortex query — all while delivering an improved customer experience,” Tengelsen says.

Realizing a gen AI future for all

These are just a few of the promising ways organizations across industries are moving their gen AI apps to production today. And with Snowflake’s built-in security and governance, bringing AI securely into your workflow has never been easier. Whether it is using Document AI or Cortex Search, Snowflake Copilot or Cortex Analyst (in public preview), Snowflake’s unified AI and data platform can help build enterprise-grade gen AI applications. 

To discover how other companies, such as Bayer and Siemens Energy, are using gen AI to increase revenue, improve productivity and better serve their customers, download Snowflake’s customer success ebook “Secrets of Gen AI Success.”

Snowflake Special Edition Generative AI and LLMs for dummies: Embrace generative AI and LLMs with the Snowflake Data Cloud

Sports organizations that have prioritized modernizing their tech stack are now in a position to leverage their data to build a competitive edge using AI and ML.  

The bar has been raised in terms of fan expectations for attending live sporting events.  Attending a game live must deliver significant value for fans beyond the more convenient options of streaming a game on your mobile device or watching from your couch. More entities will use AI on their fan data to make attending games both more convenient and personally gratifying. This could take the form of more facial-recognition-enabled entry options, mobile alert updates on lines for bathrooms near your seating section, customized parking options and sponsor-enabled personalized merchandise and concession offers.  

Sportsbooks are already required to adhere to a robust set of compliance and reporting requirements while also balancing their profit and loss. Part of the complexity is based on varying rules by country and US state.  

As problem gambling remains a rising concern in more mature sports betting markets such as Europe and younger markets such as the United States, more federal and state guardrails will likely be rolled out — with more scrutinous auditing to police against potential bad actors. More sportsbooks will lean on AI to automate processes for managing these requirements and on machine learning to maximize revenue opportunities with minimal risk exposure.  

Leagues, federations and teams now have efficient solutions for ingesting data from any source, regardless of structure. By leveraging the data from more tracking devices, motion sensors, wearables, videos and other sources, they can strengthen their competitive edge. Likewise, sportsbooks require unique data sources to differentiate their betting products from those of competitors.

With a robust data strategy in place, sports organizations will have confidence in the quality and security of their data and lean more on AI to extract insights from past performances and outcomes and on ML to predict roster and revenue impact. 

Read the Snowflake AI + Data Predictions 2025 report for insights from seven other industry leaders and the latest big-picture data and AI forecasts from leaders such as Snowflake’s Head of AI,  Baris Gultekin, and Chief Information Security Officer, Brad Jones.

In this new era, marketers and advertisers hold the power. Snowflake’s open ecosystem lets them harness best-in-class solutions and build tech stacks that revolve around their data — not the other way around. Centered on business models that allow you to pay for what you use, efficiency and agility are no longer just aspirational trade-offs for effective cost management; they’re the new reality. As marketers must do more with less, the future belongs to those who embrace this new paradigm. The revolution has arrived, and marketers are poised to lead the charge by leaning into data and marketing platforms that harness this momentum.

Snowflake and The Trade Desk — a collaboration powering data-driven advertising

At the heart of this convergence lies the collaboration between Snowflake and The Trade Desk, which is a game changer for transparent, data-driven advertising. Together, we’re empowering brands to tap into the full potential of their first-party data — seamlessly and with privacy at the forefront. Imagine being able to segment and refine your audience targeting, simplify your first-party data activation workflows, and get clear insights into your campaign performance — all while working with best-in-class technologies. That’s what Snowflake and The Trade Desk are committed to doing, together. Here are a few of our joint offerings that advertisers can use to power real-time first-party data flows, enhancing campaign effectiveness and measurement.

Plan: Unified ID 2.0 (UID2) to enrich and segment

Unified ID 2.0 helps advertisers enrich their first-party data and precisely segment their audience. By leveraging a persistent key and addressability and enrichment solution, advertisers can drive 1:1 targeting and measurement from a single location. They can also convert directly identifying information to UID2s in Snowflake without moving data. Enriching first-party data helps advertisers enhance personalization and segmentation while maintaining privacy compliance. This solution also helps to improve addressability by unifying user identity across channels for accurate targeting without third-party cookies.

Activate: First-party data

Advertisers can prepare first-party data for campaign activation on The Trade Desk, push customer data to The Trade Desk, and activate audiences without needing to build directly to an API or paying a 3rd party provider. This lets them deliver targeted ads to re-engage customers already interested in a brand and create lookalike models to discover new prospects similar to core customers.

Measure and optimize: Conversion API (CAPI) Native App 

Coming soon to Snowflake Marketplace is a solution that allows advertisers to capture online and offline conversion events in The Trade Desk to drive understanding of in-flight campaigns. With this solution, advertisers can upload conversion events without needing to build directly to an API or paying a third-party provider. This helps them improve attribution accuracy and allocate budget more effectively across channels and enhance campaign relevance by targeting the users most likely to convert.

Measure and optimize: Raw event data stream (REDS)

REDS enables log-level analysis of multi-channel campaigns for deeper segmentation and eliminates data ingestion, reducing consumption and storage costs. This solution helps marketers build multi-touch and custom attribution models to optimize ad spend. It also lets marketers create advanced audiences directly from Snowflake in order to target customers with personalized messaging.

The growing momentum of martech integrations with The Trade Desk through Snowflake

As marketers increasingly rely on first-party data, customer data platforms (CDPs) are integrating with The Trade Desk through Snowflake in unprecedented ways. Simon Data, Amplitude, Tealium and Twilio Segment are just a few examples of CDPs that are leveraging this integration to provide marketers with concentrated, real-time first-party data flows, enhancing campaign effectiveness and measurement. 

This ongoing transformation is not a one-off event — it's a journey that continues to gain momentum. Simon Data has been at the forefront of this integration, building an impressive track record with The Trade Desk. What began as a collaboration with a challenger brand is now rapidly scaling, with seven clients already live as of October 2024. Simon's goal is to have more than 75% of their clients using the integration by the second half of the year. The feedback has been overwhelmingly positive, and performance speaks for itself. With a match rate of 60% for most clients — and an additional 20% lift when using Simon Data’s Match+ product — the integration is helping advertisers do more with the data they already have.

These integrations mark a significant advantage for the providers involved: multiple platforms to partner with to better serve marketers and a growing concentration of first-party data are driving improved targeting, measurement and reach. And the big winners here? That would have to be marketers and advertisers, who now have a powerful set of tools that allow them to navigate the complexities of privacy while still delivering relevant, timely content to their audiences. One standout benefit is the ability to run daily segment syncs with The Trade Desk, keeping audiences fresh and engagement-ready, a significant improvement over traditional weekly sync cycles. This freshness can directly translate into better performance for advertisers and a stronger, more nimble approach to audience targeting.

CDP leaders weigh in

Martech providers such as Treasure Data, Amplitude, Tealium and Segment have embraced this integration wholeheartedly. Here's what they have to say about the benefits:

Treasure Data: “The integration of Snowflake and Treasure Data, along with Snowflake’s frictionless data sharing integrations with platforms such as The Trade Desk, empowers advertisers with rich customer insights to fuel personalized campaigns across channels. With customer profiles and insights updated in real time within Treasure Data and seamlessly connected to Snowflake, personalized engagement remains relevant, driving optimal outcomes and maximizing ROI.”
—Kaz Ohta, CEO and Co-Founder, Treasure Data

Twilio Segment: "Marketers can deliver the personalized experiences consumers expect by activating first-party audiences with Segment, Snowflake and The Trade Desk — all while minimizing data movement. As insights flow back through Snowflake into Segment, marketers can optimize attribution, refine targeting and tap into new opportunities for targeting on The Trade Desk platform.”
—Kevin Harris, VP of Partnerships, Twilio Segment

Tealium: “With the convergence of martech and adtech driven by the market's shift to first-party data, enterprises face increasing demands to manage and activate data across systems, while ensuring compliance and performance at scale. Tealium excels at collecting, enriching and unifying first-party data in real time, creating robust customer profiles. By integrating this enriched data with Snowflake’s platform via Snowpipe for continuous ingestion, enterprises can automate data flows and seamlessly leverage Snowflake and Tealium to send audiences to activation platforms, like The Trade Desk, without duplicating data. This streamlined process enables enterprise clients to deliver highly personalized, efficient advertising campaigns, meeting the complex needs of large-scale operations.”
—Matt Gray, Global VP of Partnerships, Tealium

Amplitude: “By integrating Amplitude with Snowflake, a brand’s behavioral cohorts are enriched with additional transactional, identity and contextual information throughout the omnichannel. Snowflake’s real-time data transfer with The Trade Desk then enables advertisers to confidently run their multi-channel campaigns. Measured value across a proven network of publishers in retail media, news sites, social and hundreds of other meaningful media destinations is then activated, leading to more impressions, higher conversion rates and most importantly, reliable retention.” 
—Ted Sfikas, Field CTO, Amplitude

These integrations allow marketers to lean into their strengths, let complementary partners handle the heavy lifting and create a streamlined, effective workflow that serves both technical and nontechnical users alike.

Looking ahead

The convergence of martech and adtech around first-party data is just the beginning. As Snowflake and The Trade Desk continue to deepen their collaboration, the ecosystem is evolving into one that is more open, interoperable, and responsive to the changing needs of businesses and consumers alike. This trend is set to unlock even more opportunities, helping marketers, advertisers and data scientists succeed in a privacy-centric, data-driven world. 

The future is bright for those who lean into this convergence and take advantage of best-in-class solutions that bring the technology to their data and empower them to maximize its value.

Want to get started with the solutions highlighted above? Visit Snowflake Marketplace:

• Try First-party Data Activation to onboard your first-party data for campaign activation.

• Leverage REDS data today on Snowflake to build custom attribution models and advanced audience segmentation tailored to your business needs.  

• Start using deterministic identifiers like email or phone numbers with built-in transparency and privacy controls with Unified ID 2.0.

Discover how leading brands and agencies are leveraging REDS data to drive smarter marketing strategies. Register for the webinar on December 12 with Horizon Media and The Trade Desk. 

As we come to the end of a calendar year, it’s natural to contemplate what the new year will hold for us. It’s an understatement to say that the future is very hard to predict, but it’s possible to both prepare for the likeliest outcomes and stay ready to adapt to the unexpected. 

In the enterprise technology space, both the greatest certainties and the most significant potential surprises come from one area: the rapidly advancing field of artificial intelligence. Thus, as we consider 2025 and beyond, it’s important to focus a lot of attention on the development and adoption of AI.

Together with a dozen experts and leaders at Snowflake, I have done exactly that, and today we debut the result: the “Snowflake Data + AI Predictions 2024” report. Along with issues of AI advancement, we considered directional trends and urgent needs in cybersecurity, open source software and more, but quite naturally a lot of our conversations turned to AI and how this fast-moving, volatile area of technology may continue to surprise the world.

2025 will be the year that many enterprises move from experimenting with LLMs and generative AI to operationalizing them, which will bring its own challenges. From my perspective, these are the key ideas that emerged from our discussions of AI and particularly its impact on the enterprise.

• AI observability is essential to operationalizing AI, and platforms will roll out solutions. When you’re running a large language model, you need observability into how the model may change as it ingests new data. It’s also important to have visibility into cost and performance. AI observability solutions are emerging to meet this need, but over time it’s most likely that the large data platforms, including Snowflake, will provide the solutions.

• Hallucinations will slow the rollout of customer-facing AI. The models keep getting better, and techniques such as retrieval augmented generation (RAG) will help reduce hallucinations and errors and put up guardrails that protect sensitive data and the voice and tone of a company. But businesses will continue to hesitate to put in front of customers a technology that may display bias or provide inaccurate responses. For this reason, internal-facing AI will continue to be the focus for the next couple of years.

• The next evolution in data is making it AI ready. For years, an essential tenet of digital transformation has been to make data accessible, to break down silos so that the enterprise can draw value from all of its data. This remains important, of course, but the next step will be to make sure that the enterprise’s unified data is AI ready, able to be plugged into existing agents and applications. 

  The trend to centralize data will accelerate, making sure that data is high-quality, accurate and well managed. Beyond working with well-structured data in a data warehouse, modern AI systems can use deep learning and natural language processing to work effectively with unstructured and semi-structured data in data lakes and lakehouses. Overall, data must be easily accessible to AI systems, with clear metadata management and a focus on relevance and timeliness. And data strategy must evolve to make sure that AI initiatives are aligned with business goals and are effectively instilling a data-driven culture in the organization.

• Expect autonomous agents, document digestion and AI as its own killer app. Our report notes that LLMs and generative AI will be so deeply embedded into how we live and work that thinking of a “killer app for AI” is like thinking of a killer app for electricity. But if we’re looking for the short-term winner, it’s going to be internal-facing use cases that let workers pull insights from massive troves of unstructured data. Snowflake recently helped a customer ingest about 700,000 pages of research and make it easily consumable through a conversational chatbot, allowing analysts to glean insights that had been functionally unavailable, though the company had the data. That will remain a major use of generative AI for some time.

  But in the next few years, the game-changing breakthrough in how we work with AI will be autonomous agents. Rather than answering a specific question, independent agents will act on broad instructions from a human user. “Create and launch a marketing campaign to attract this key customer cohort” could be automatically broken into subtasks such as designing on-brand copy graphics, making ad buys to reach the desired audience and optimizing based on initial performance.

• Leadership will be the antidote to AI exhaustion. AI has been advancing so quickly that the project that consumed a team’s every waking hour two weeks ago could be completely outdated tomorrow. Do you move forward or redo the work? If the latter, what if it happens again next week? Everyone I know in the AI space has talked about burnout at some point in the past year. To keep teams at their productive, creative best, leaders need to step in. We must set our sights on goals and ROI, rather than focusing on the shiny object. AI projects should not be about “the latest” or “the best.” Like any business decision or investment, we must weigh what’s most effective in terms of results and resources.

These thoughts are just some of what’s in the report. At the societal level, we look at the interplay of industry guardrails and regulatory oversight. Our cybersecurity experts tackle the ways that AI will both empower attackers and provide new ways to fight them. We look at developments in open source technologies that will allow organizations to improve their data strategies. And we talk about how leaders can keep up with the sometimes unnerving pace of change. Check out “Snowflake Data + AI Predictions 2025” for the whole story.

How to navigate such a landscape is top of mind for me and top executives such as Snowflake’s CEO, Sridhar Ramaswamy, Snowflake’s Distinguished AI Engineer, Yuxiong Xe, and other industry-specific experts who weighed in on the Snowflake AI + Data Predictions 2025 report. From operationalizing AI to rewriting your leadership skill set, we’re predicting what an AI-accelerated future might look like (and what might happen if organizations don’t get their foundational data strategies in place to be part of it). 

We dive into how AI adoption and anticipated regulatory challenges will affect financial services in years to come. The industry’s approach will be more measured, balancing innovation with clear, demonstrable business value. One of the big issues is how willing the tech-forward financial industry will be to bet on very new, entirely unproven AI solutions.

For the rest of the financial services predictions and more, download the report Snowflake AI + Data Predictions 2025.

I see AI adding value in two ways across financial services organizations: augmenting workers and operating autonomously without human involvement. The latter will remain in the experimentation stage for some time with too high a risk of negative customer outcomes hurting the bottom line. 

The board is going to care about ROI, of course, but not without safeguarding the organization. Naturally, regulators are also vigilant about protecting firms, consumers and the financial system — hampering any ultra-fast-paced rollouts organizations might have in the works. 

Read the full report for the rest of my 2025 financial services predictions, insights from six other industry leaders and the latest big-picture data and AI forecasts from leaders such as Baris Gultekin, Snowflake’s Head of AI, and Brad Jones, Snowflake’s Chief Information Security Officer. 

This enhanced level of protection adds to the growing security capabilities of Snowflake Horizon Catalog, which empowers security admins and chief information security officers to better safeguard their security posture and mitigate risks of credential theft. It also follows our previous announcement that multi-factor authentication (MFA) will be the default for all password sign-ins in new Snowflake accounts created starting October 2024.

In order to hit this milestone, and to make sure we build a viable path for all customers to migrate, we are taking a phased approach. Before we expand on the phases, let’s lay down some taxonomy:

• Account: This refers to the container that contains various objects, such as tables, views, databases, schema and user accounts. See here for more information. 

• Users: This refers to objects that show the identities of people who can access customers’ accounts and the objects within them. See here for more information.

• Human users: This refers to users who are human and normally use an interactive login to sign in to Snowflake. Such users are declared in the Snowflake user object with TYPE = PERSON or NULL (by default is NULL). See here for more information.

• Service users: This refers to users that are used for programmatic access without interactive login. Such users are declared in the Snowflake user object with TYPE = SERVICE or LEGACY_SERVICE. Neither SERVICE nor LEGACY_SERVICE will be subject to Snowflake MFA policies. SERVICE users cannot use passwords to sign in. LEGACY_SERVICE is meant for applications that take longer to update and move away from passwords; as such, LEGACY_SERVICE has a temporary exception to use passwords until the app is updated. See here for more information.

Our phased approach will entail three stages:

1. April 2025: Enable for all accounts the default authentication policy, with MFA enforced on password sign-ins for human users. In this phase, all human users in accounts without a custom authentication policy will be required to enroll in MFA upon their next password-based sign-in to Snowflake. If an account already has a custom authentication policy at the time of this rollout, human users will not see a difference in their sign-in experience. At this time, we will also block access to Snowsight for LEGACY_SERVICE users.

2. August 2025: Enforce MFA on all password-based sign-ins for human users. In this phase, even if the customer has a custom authentication policy already defined, all human users will be required to use MFA when signing in with passwords. 

3. November 2025: Block sign-in to Snowflake using single-factor authentication with passwords for all users (human or service). In this phase, LEGACY_SERVICE is deprecated and all LEGACY_SERVICE users will be migrated to SERVICE users.

Note that these policies have no bearing on single sign-on users (using SAML or OAuth) or users using key-pair authentication. 

To help with migrations, we have created a white paper and an accompanying video migration guide.  We also added a new scanner package to the Trust Center called Threat Intelligence (generally available) that can quickly scan your account and find users that are at the risk of losing access (see here for step by step guidance). We have also been working with our partners and ecosystems — including Tableau — to help prepare their solutions for our vision for stronger authentication.

Snowflake will continue investing in the security capabilities of our customer accounts and bring more products and innovations to this space, such as native support for passkeys and time-based one-time password (TOTP) including authenticator apps. These will all work hand-in-hand with Snowflake’s other recently announced capabilities, including Leaked Password Protection, Trust Center, MFA policies, Programmatic Access Tokens (private preview soon) and many more. Stay tuned for updates!

Forward Looking Statements

This article contains forward-looking statements, including about our future product offerings, and are not commitments to deliver any product offerings. Actual results and offerings may differ and are subject to known and unknown risk and uncertainties. See our latest 10-Q for more information.

For organizations considering moving from a legacy data warehouse to Snowflake, looking to learn more about how the AI Data Cloud can support legacy Hadoop use cases, or assessing new options if your current cloud data warehouse just isn’t scaling anymore, it helps to see how others have done it. 

That’s why we’ve collected these migration success stories to help you get started on your migration to Snowflake. This blog post is the second in a three-part series on migrations. Today we’re focusing on customers who migrated from a cloud data warehouse to Snowflake and some of the benefits they saw.

A consolidated data system to accommodate a big(ger) WHOOP

When a company experiences exponential growth over a short period, it’s easy for its data foundation to feel a bit like it was built on the fly. Data becomes distributed across multiple platforms; different teams end up using different tools. They watch costs skyrocket while performance degrades. 

In essence, that was the story of WHOOP, the Boston-based wearable technology company aimed at enhancing human performance and endorsed by superstar athletes such as LeBron James and Cristiano Ronaldo. Before it migrated to Snowflake in 2022, WHOOP was using a catalog of tools — Amazon Redshift for SQL queries and BI tooling, Dremio for a data lake, PostgreSQL databases and others — that had ultimately become expensive to manage and difficult to maintain, let alone scale. 

In Snowflake, WHOOP found a simplified, fully managed platform with near-unlimited scalability and strong governance controls — in short, everything its previous system lacked. And the availability of a large network of partners who offer solutions to a whole range of problems has been an invaluable asset. The move itself took just a matter of three months, including the time it took to clean up and organize much of its existing data to set WHOOP up for the future. 

Now, the company is enjoying the benefits of Snowflake’s performance, simplicity and data governance. With separated compute across warehouses, there’s no longer any worry about one team’s queries straining another’s resources, and features such as Iceberg Tables are simplifying pipelines and saving the company money. 

Nexon finds 7x performance improvements by unifying its massive data stores

Founded in 1994, Nexon is a company engaged in the production, development and operation of online games and Virtual Worlds. Serving a company that has games available in more than 190 countries and employs more than 8,000 people, its data engineering team is always running. Processing some 90,000 tables per day, the team oversees the ingestion of more than 100 terabytes of data from upward of 8,500 events daily. So when the company sought to unify all of its data on a single platform, it knew it needed something scalable, reliable, secure and convenient. 

Moving its data warehousing workload from a cloud data warehouse to Snowflake and from legacy Spark to Snowpark, Nexon saw a sevenfold performance improvement, translating to $4.5 million in cost savings annually. With an internal user base of 2,000 — and growing — the company particularly appreciated the seamless data access controls and the ability to securely share data with just a few simple clicks. With Snowpark, Nexon found processing speeds to be equally fast but more convenient and cost-effective since data never has to move off of Snowflake. 

A simpler, better solution 

While a desire to consolidate and make sense of their data foundations drove the move to Snowflake for both WHOOP and Nexon, the companies also enjoyed immediate improvements in performance, ease of use and cost. Snowflake’s deep well of partners and the growing number of tools that integrate seamlessly with its platform offer the flexibility to tackle each customer’s unique challenges.

To learn more about migrating and modernizing your legacy data platform, visit our Migrate to the AI Data Cloud web page or read more customer stories here to find out how companies such as Lucid, Big Fish Games and Business Insider have found success with Snowflake.

One key to successfully leveraging gen AI models is the ability to share data. Companies with valuable data that can be used to fine-tune LLMs want to be able to monetize it and use it for fine-tuning without granting access to the original data sources. They also want to ensure that all usage is appropriately attributed back to them. 

Unfortunately, many of the currently available solutions do not give enterprises the tools to share data safely and securely while:

• Ensuring that an organization’s valuable data is always managed by that organization and not made available to other parties, which may result in inappropriate or possibly malicious use

• Ensuring that third-party models used within the enterprise are safely sandboxed

• Carefully monitoring access to data and models

At Snowflake, we are tackling these challenges head-on and making it easier for developers to deliver trusted AI with enterprise data.

At our recent BUILD 2024 dev conference, we highlighted three features to help you share your fine-tuned LLMs, share data sets to train your LLMs, and share traditional AI/ML models safely and securely both within and outside your organization across the AI Data Cloud. We provided an overview of these features in a previous blog post, but now let’s take a closer look at how you can put them to work in your projects.

Share Snowflake Cortex AI fine-tuned LLMs from Meta and Mistral AI

To fully leverage foundational AI models, enterprises need to customize and fine-tune them to their specific domains and data sets. This task generally comes with two mandates: no data leaves their premises at any time, and no heavy investments are made in building infrastructure. 

Snowflake now offers enterprises the ability to fine-tune leading models from Meta and Mistral AI using data within their own security perimeter and without having to manage any infrastructure. Better yet, developers can easily govern and manage their custom LLMs with Snowflake Model Registry.

With Secure Model Sharing (currently in public preview), you can fine-tune and share custom foundation models in three steps:

1. Select the base model and provide your training data set as part of the FINETUNE function or by using the no-code experience in Snowflake AI & ML Studio. The fine-tuned models can be used through the COMPLETE function.

2. Securely share your fine-tuned models with other Snowflake accounts in your region.

3. Replicate your fine-tuned models across regions within your organization.

Unlock the power of Cortex LLMs with Cortex Knowledge Extensions

Enterprises want an easy way to augment their foundation models with domain-specific information to make them provide more relevant responses. Traditionally, it takes a lot of time and effort to find and procure the right data sets, and then more time and technical skill to prepare the data for consumption and fine-tune the LLMs. Snowflake has already streamlined the first part of that process — locating appropriate data — with Snowflake Marketplace, which offers one central location to quickly find, try and buy more than 2,900 data sets, apps and data products (as of October 31, 2024). Now, with Cortex Knowledge Extensions (currently in private preview), we’re making it easier to prepare and transform third-party data.

Cortex Knowledge Extensions give customers an “easy button” for augmenting their chosen foundation model with up-to-date information in a particular domain without requiring additional technical expertise to fine-tune and massage raw data from a content provider. Critically, customers will have the confidence that they are using officially licensed content.

Cortex Knowledge Extensions allow gen AI applications to draw responses from providers' unstructured, licensed data while giving them appropriate attribution and isolating the original full dataset from exposure. This helps providers monetize and participate in gen AI while minimizing the risk of their content being used for model training purposes. 

To make their data available on Snowflake Marketplace, the content provider sets up a Cortex Search service on their data and publishes to Snowflake Marketplace. Once published, a consumer can find the listing and acquire the data from Snowflake Marketplace. Consumers can then use Cortex AI APIs to prompt LLMs with the acquired Snowflake Marketplace data.

Share traditional AI/ML models in the AI Data Cloud

More and more enterprises are building custom AI/ML models for specific tasks such as predicting churn or forecasting revenues. These models may be developed within the organization by data scientists or externally by partners. Enterprises can now unlock the power of these models and share them with partners, customers and users within the enterprise using Snowflake Native Apps on both Internal Marketplace and external-facing Snowflake Marketplace. 

With Snowflake Secure Data Sharing, organizations can allow end users to run ML models securely within fine-grained role-based access control on their data. The data itself never leaves the organization’s security boundary. Packaging the models with Snowflake Native Apps ensures that they inherit Snowflake Native Apps’ security posture, including security scanning, sandboxing and access to local or external resources based on specific privileges granted to the model.

Sharing a model is as simple as adding model artifacts to an application package and granting application-specific consumer usage privileges. Consumers are then free to install the application and invoke model functions.

With Snowflake collaboration and data sharing, enterprises can easily create and share AI/ML models — both traditional models and fine-tuned LLMs — and share their benefits with the rest of the enterprise. To learn more and try out some of these features, check out these resources: 

• Secure data sharing overview

• Quickstart: Collaboration in Snowflake

• BUILD 2024 “What’s New: Snowflake Horizon Catalog” on-demand session, including Internal Marketplace

This crisis is a huge challenge for already taxed government organizations. Where should limited resources be deployed? Real-time climate data can help predict areas where conditions will be most severe and the most likely migration routes. 

These gaps are everywhere. Data disparities underpin not only climate-related challenges but also the crises we face in public health and social justice. The twin revolutions in cloud technology and artificial intelligence are producing more data and analysis than ever before. As we grow ever more reliant on data, we must ensure that data represents everyone. People can benefit from data-driven innovation only if the data sets that address their most pressing issues are accessible and include them. 

What is a data disparity? Think of systematic undertesting of women in clinical trials, unequal access to primary school education and faulty climate models. These are real problems with real impacts that stem from how we manage information. 

Solving data disparities is one of the great opportunities of our time. 

The United Kingdom experiences 32,000 excess deaths each winter. Of those deaths, 9,700 are attributable to living in a cold home — about the same as the number of people who die from breast or prostate cancer each year — and 3,200 are directly linked to people who cannot afford fuel to make a warm home. Energy insecurity should not mean life or death. 

In an ideal world, governments and nongovernmental organizations could couple their weather data with usage metrics from energy suppliers to anticipate which homes will be most vulnerable in winter and offer aid. Accessing and deploying that data has long been near-impossible, but that’s changing. EDF, which supplies gas and electricity to homes across Britain, now uses machine learning to identify financially vulnerable customers and step in to provide assistance in times of need. 

Inequitable data access exacerbates global inequalities. We now rely on data to inform the most pressing socioeconomic conversations and influence policy, but we must make a concerted effort — across private and public entities — to make that data whole. That means dismantling data silos, bridging gaps in data collection and safely and securely sharing knowledge. 

Despite commendable efforts, we’re falling short. Globally, we are on track to achieve only 15% of the United Nations Sustainable Development Goals (SDGs), which aim to secure “peace and prosperity for people and the planet.” These are the most important representations of what we must do to ensure a good life for all. More needs to be done. 

SDG 17 focuses on building partnerships between organizations for more sustainable development. This goal recognizes that no one company or organization can solve our global problems; instead, we need a global movement. 

Many of those players will be private companies. There have been notable efforts through “data for good” efforts from Meta, Google and others to provide massive data sets for use in research and studies. A new End Data Disparity campaign, which brings together data leaders like Snowflake with on-the-ground groups like the UN’s International Organization for Migration, builds on these efforts by encouraging companies to share their growing technology capabilities with those who traditionally have done with less. 

Even a leading nonprofit likely has only a couple of people on its data science team. Compare that to hundreds of data scientists at an equivalent multinational company. Nonprofits and NGOs don’t always have the expertise or capacity to fill in gaps in data sets and make complex multi-factor calculations. But that can change. 

We can draw on recent breakthroughs in AI and classic machine learning to inform how we address the world’s biggest issues. Imagine it were possible to adjust the deployment of doctors and relief workers in real time based on anonymized phone data. This could improve access to services and ultimately save lives. By sharing knowledge — listening closely to people on the ground — and using technology, organizations can work toward meeting the UN’s Global Goals.

As world leaders convene for crucial discussions at the United Nations General Assembly, COP29 and the World Economic Summit Davos, let’s make sure to keep data on the agenda. 

It’s time we end data disparity together.

Enter the Composable CDP on the Snowflake AI Data Cloud for Travel and Hospitality. Trusted by leading brands, this innovative solution empowers travel and hospitality companies to unlock the full potential of their data in Snowflake, helping deliver tailored guest experiences and optimizing loyalty programs.

How Composable CDPs work on the Snowflake AI Data Cloud

The Snowflake AI Data Cloud centralizes customer data from reservations, loyalty programs, booking engines and more. A Composable CDP taps into this unified source to provide travel and hospitality brands with the tools to activate their data for marketing, loyalty management and guest personalization. 

Key components of the Composable CDP for travel and hospitality include:

• Data activation: Build audiences based on guest profiles, preferences and behaviors. Sync audience and guest data to email platforms, customer relationship management (CRM) systems, advertising platforms or any other marketing tool that drives personalized travel experiences.

• AI Decisioning: Use AI to continuously optimize marketing efforts with tailored promotions, offers and loyalty communications for each guest to enhance their experience and maximize performance.

• Campaign intelligence: Analyze marketing campaigns across platforms, and identify opportunities to improve guest experiences. Actionable insights can be turned into new audience segments or refined communication strategies that drive better-performing marketing and guest engagement.

• Event tracking: Capture guest interactions across websites, apps and on-property activities, and store this data directly in Snowflake.

• Identity resolution: Create unified guest profiles that combine data from multiple sources (e.g., reservations, loyalty accounts and on-property interactions) to provide a comprehensive view of each guest.

By empowering travel and hospitality brands to act on all of their data, a Composable CDP enables personalized experiences that drive guest loyalty and satisfaction.

Why Composable CDPs suit travel and hospitality companies

The travel and hospitality sector demands a flexible, secure data solution that can handle the complexities of managing guest interactions across multiple brands, channels and properties. Composable CDPs operate on data where it already lives — within the company’s Snowflake AI Data Cloud — which helps you ensure that data remains secure but also easily accessible for marketing teams.

Key benefits:

• Seamless integration: A Composable CDP works directly with existing data in Snowflake, meaning no data duplication or complex transformations are necessary. This is critical for travel and hospitality businesses managing data created by multiple systems, including property management systems, loyalty platforms and booking engines.

• Flexible data models: Every travel brand is unique. Whether the data is structured around properties, trips or loyalty programs, a Composable CDP adapts to the data schema, enabling businesses to act quickly on their comprehensive data.

• Customization for multibrand and multinational companies: Composable CDPs offer flexible tooling so that conglomerates, with brands operating across multiple regions, can partition data access and usage. This makes it easy for all team members, regardless of division, to leverage the same data in Snowflake and only act on the subset that is relevant to their brand, region or use case. 

• Easy AI and machine learning integration: Travel brands can use Snowflake Cortex AI to build and deploy machine learning models to predict guest behavior, optimize offers and improve the overall guest experience. With AI Decisioning, marketers can optimize lifecycle marketing campaigns and target each guest with hyperpersonalized messages that drive maximum engagement.

• Trusted data governance and security controls: Travel brands can leverage Snowflake capabilities to help comply with data regulations. A Composable CDP benefits from Snowflake’s built-in governance to help customers manage how data is accessed.

Proof points: Travel and hospitality leaders driving results with the Composable CDP

Leading multi-brand multi-national hotel chain

• Who they are: One of the world’s leading hospitality companies, with a portfolio of more than 5,000 properties globally

• Whom they serve: Guests across a wide range of hotels, from luxury brands to budget accommodations.

• How they’re using a Composable CDP on the AI Data Cloud for Travel and Hospitality: The company leverages a Composable CDP to enhance loyalty programs, offering personalized promotions and recommendations to frequent travelers.

• Outcome: They improved guest retention by personalizing guest experiences across properties, resulting in higher loyalty program participation.

Leading gaming and hospitality company

• Who they are: A leader in the casino and entertainment industry, offering a wide range of hospitality and entertainment experiences across multiple properties.

• Whom they serve: Individual guests and loyalty members who visit casinos, hotels and resorts.

• How they’re using the Composable CDP on the AI Data Cloud for Travel and Hospitality: The company uses a Composable CDP to unify guest data across properties and personalize marketing offers for loyalty program members.

• Outcome: It saw an increase in engagement with personalized offers and loyalty communications by activating guest data across channels through a Composable CDP.

Transform your travel and hospitality business with the AI Data Cloud

Composable CDPs, powered by the AI Data Cloud for Travel and Hospitality, enable brands to deliver personalized guest experiences, optimize marketing campaigns and drive loyalty — all while maintaining strict data governance and compliance settings. With data stored securely in Snowflake, travel brands can confidently act on their data and unlock its full power.

To learn more about how you can transform your business with a Composable CDP on Snowflake, explore the Snowflake AI Data Cloud for Travel and Hospitality. Additionally, if you plan to be at Phocuswright 2024 in Phoenix, please reach out to whitnee.hawthorne@snowflake.com to book a meeting.

This article highlights nine key areas to watch out for and plan around in order to accelerate a smooth transition to the cloud. Additionally, this blog will shed light on some of Snowflake's proven features to help you optimize the value of your migration efforts.

Migrating enterprise data to the cloud can be a daunting task. However, when executed properly, it can be both efficient and far less challenging. Leveraging Snowflake’s built-in features can further alleviate some of the common pain points associated with the migration process.

The areas of focus in this article are:

1. Data compression

2. Initial data uploads 

3. Ongoing data uploads

4. Data set prioritization

5. Data lifecycle management

6. Data security and encryption

7. Data validation

8. Disaster recovery 

9. Multiple software environments

Data compression

Data compression is crucial for conserving bandwidth when transferring data from on-premises to the cloud. There are several ways to compress data before uploading it. For instance, gzip is a reliable compression method. When loading data into Snowflake from Amazon S3 buckets, data compression can optimize the process, improving efficiency and reducing transfer time.

How Snowflake can help: If files are compressed using gzip or another widely used format, Snowflake can directly ingest the compressed data without requiring manual decompression. Alternatively, if your files are uncompressed on a local drive, Snowflake will automatically compress them using gzip — unless compression is explicitly disabled or a different compression method is specified. This built-in feature further helps conserve bandwidth during file uploads, making the migration process more efficient.

Initial data uploads

Every enterprise manages vast amounts of data spread across different formats in on-premises systems. A hybrid approach, where some data sets remain on-premises and some are moved to the cloud, may seem appealing to ease the upfront burden, but that will likely be much more challenging to manage long-term. With a hybrid approach, you are tasked with managing two separate sets of infrastructure, potentially different formats, and a federated model is likely time-consuming and expensive to use.

Data size can range from a few gigabytes to multiple terabytes. Handling a few gigabytes (GBs) is relatively straightforward, but migrating data in the terabyte range can pose logistical challenges. To help ensure success of this massive undertaking, a one-time, tamper-proof transfer method is essential to promote data accuracy and maintain security controls throughout the process.

How Snowflake can help: Every major cloud service provider (CSP) offers solutions to assist with large-scale data transfers. AWS provides Snowball, Microsoft Azure offers Databox, and Google has the Transfer Appliance to facilitate one-time, massive data migrations. Since Snowflake is compatible with these CSPs, once the offline transfer is complete and the data is available in the cloud, ingesting it into Snowflake for further processing becomes a seamless process.

Ongoing data uploads

While one-time uploads can be managed using the solutions discussed above, customers must also consider how to handle new data generated on a daily basis. This process could continue indefinitely or for a fixed period until the on-premises architecture is fully retired and data is piped directly into your cloud platform. To meet these ongoing data load requirements, pipelines must be built to continuously ingest and upload newly generated data into your cloud platform, enabling a seamless and efficient flow of information during and after the migration.

How Snowflake can help: Snowflake offers a variety of options for data ingestion. For real-time, continuous loading, Snowpipe is ideal for trickle feeds. For batch loading, the powerful COPY command can be utilized. For low-latency streaming use cases, Snowpipe Streaming is ideal. Additionally, Snowflake’s robust data integration ecosystem tools enable secure and controlled incremental uploads without the need for complex infrastructure. This flexibility allows data ingestion to be efficient and reliable, with minimal disruptions during the migration process. You can learn more about data ingestion best practices with Snowflake in this three-part series: Part 1, Part 2, Part 3.

Data set prioritization

Enterprises often face the challenge of different teams competing to migrate their data to the cloud as quickly as possible. If not managed systematically, this can lead to multiple copies of the same data being stored in the cloud, creating inefficiencies. To avoid this, it's crucial to prioritize data sets and migrate them in a structured sequence, starting with "master data sets" before moving on to others.

While Snowflake facilitates seamless data migration and prioritization, many of our customers have demonstrated that thorough planning and careful identification of data sets are key to ensuring the right data is moved first, preventing unnecessary duplication. It can be as simple as listing down the data sets in a central location like Sharepoint and assigning priority to help plan appropriately and reviewing the list on a periodic basis.

How Snowflake can help: While there are numerous methods for uploading data sets and we have discussed a couple of them already in this blog, the option to load files using Snowflake's web interface stands out as one of the easiest and often the quickest way to ingest data. This user-friendly approach allows business users to swiftly transfer files into Snowflake, streamlining the data-ingestion process.

Data lifecycle management

Data lifecycle management is a critical area for effective cost management in the cloud. Maintaining data in the cloud incurs operating costs, so establishing a robust data retention policy should be a foundational aspect of a customer’s cloud strategy. While regulatory and compliance requirements may prevent complete data deletion, implementing an expiry model for data that doesn't fall under these retention requirements is recommended. This approach helps optimize storage costs.

How Snowflake can help: Snowflake offers several features that ease data lifecycle management, including various data storage considerations. These, combined with our cost optimization tools like budgets, help reduce storage costs. Additionally, our product team is working on new policy-based capabilities to make the lifecycle of data seamless to manage.

Data security and encryption

Data security is an important area that organizations consider when moving their data to the cloud. The security team has to be brought on board with the idea that enterprise data will be leaving the four walls of the enterprise and moving to the cloud. Features like private connectivity, network policies and encryption are some of the widely adopted methods for securing data during movement to the cloud.

Some organizations have established security policies that require data to be encrypted before it leaves their data center. Encryption methodologies, such as RSA and AES, can be applied at the file level to enable data protection during this process. Once the data is in transit to your cloud platform, comprehensive data protection policies can be implemented to safeguard the data both in transit and at rest, providing an additional layer of security throughout the migration process.

How Snowflake can help: Snowflake offers end-to-end encryption to help organizations meet their compliance requirements, keeping the data secure throughout its lifecycle. Additionally, Snowflake provides robust key management solutions once the data is under its management, further enhancing security and control over sensitive information. In addition, Private Link and limiting HTTP request acceptance from certain IP addresses (also known as “IP whitelisting”) help to limit data access.

Data validation

Data validation is crucial for data quality and instilling confidence in business users as they utilize this information. Some key metrics that customers commonly use for validation include the number of unique values, number of null values, data set freshness and duplicate values. Regularly logging and reviewing these metrics at defined intervals helps maintain data quality and supports informed decision-making for the business groups.

How Snowflake can help: Snowflake offers a variety of data metric functions that can run in the background to help identify anomalies and support data validation. These functions continuously monitor the data, enabling proactive detection of issues and promoting the overall quality and reliability of the data.

Disaster recovery

The level of disaster recovery (DR) preparedness required for a cloud differs significantly from an on-prem system. By default the CSPs have established standards to help DR strategies for maintaining data copies. While on-premises solutions often necessitate extensive planning and resources for data redundancy and to adhere to RPO and RTO policies for recovery, CSPs typically offer built-in DR capabilities that streamline these processes and enhance data resilience. This allows organizations to leverage the CSP’s infrastructure for more efficient and effective disaster recovery. Focusing on application needs from a data-availability standpoint helps in mitigating business risks.

How Snowflake can help: One of the key strengths of Snowflake is its capability to provide  seamless business continuity across different clouds and regions using Snowgrid, which is very easy to implement without a lot of infrastructure plumbing in the backend. In addition, Snowflake provides several built-in features to support disaster recovery, including automatic replication, time travel, failover/failback and secure data sharing . 

Multiple software environments

In the cloud, the need for multiple environments (such as development, testing, staging and production) often persists, similar to on-premises setups. However, cloud platforms offer greater flexibility and scalability which can simplify management. One can save on costs as the cloud allows for on-demand allocation of resources, helping enterprises stand up and tear down environments as needed and paying only for what they use. In addition, automation tools for deployment and maintenance of the environments make it a breeze to manage all of the logistics. User testing, performance testing, regression testing, security testing and more become very easy due to the nature of the cloud.

How Snowflake can help: Snowflake helps enterprises to save time, effort and money by providing a centralized platform for easy access, zero copy cloning for instant copies without replication across environments, integration with CI/CD tools and instant access to resources to help with different types of testing without the added management of maintaining the infrastructure needed to support these capabilities.

Closing thoughts

While we have discussed the nine broad areas where we have seen customers struggle and the potential solutions, this is by no means an exhaustive list. With careful planning and the right tools, migrating enterprise data to the cloud can make a cumbersome task easy to plan and manage. Snowflake’s robust set of features, ranging from data compression, upload options, data lifecycle management and enhanced security, help accelerate that journey to the cloud while minimizing risks.

By focusing on the critical areas discussed in this article, organizations can optimize their cloud migration efforts, ensuring a smooth transition that aligns with both operational needs and long-term business goals. With Snowflake as a trusted partner by your side, the journey of your enterprise data to the cloud is smooth.  For further reading, please visit Snowflake’s dedicated migration page, Migrate to the Cloud, and learn more about our native code-conversion tooling, SnowConvert. 

In these conversations, there are a number of questions that I hear time and time again: Will my data platform be scalable and reliable enough? Will it be easy to use for my entire team? What will costs look like? How will my data stay secure and governed? 

A critical part of this decision is determining which foundational technology to build infrastructure on. Managed Apache Spark environments — such as Databricks, Amazon EMR, and certain Cloudera deployments — can present teams with a plethora of pain points, which may include complexity, unpredictable costs, security concerns, or performance issues. 

I see these factors as key reasons why organizations of all sizes and industries make the move to Snowflake. Helping organizations through a migration — and seeing the tremendous outcomes they achieve as a result — are some of the most rewarding parts of my job. And in the new book “Secrets of Apache Spark to Snowflake Migration Success,” we’re spotlighting some of these exciting stories from customers as varied as AMN Healthcare, IGS Energy, Intercontinental Exchange and the New York Stock Exchange. 

Here are just a few examples of leading organizations that are migrating from managed Spark environments to Snowflake to save millions of dollars, improve performance and get products to market faster to delight their customers sooner. 

Travelpass delivers more curated traveler experiences — while saving 65% in costs

Built on the idea of discovering common ground through exploration, Travelpass connects travelers with the best hotels and travel experiences to meet their needs. Data is the lifeblood of Travelpass’ business — yet the Travelpass data teams were spending a lot of time focusing on how to build, rather than what to build.

By moving from Databricks to Snowflake, Travelpass now empowers more people to work with data to deliver greater efficiency, more informed decision-making and a more tailored experience for travelers across the globe. Thanks to Snowflake’s ease of use and approachability, even non-data engineers at Travelpass now contribute to Snowflake data in a meaningful, quick way.

Benefits include: 

• 65% cost savings by switching from its previous platform, Databricks, to Snowflake

• 350% improved efficiency delivering data to business units, thanks to Snowflake Dynamic Tables

• Greater reliability and ecosystem stability by eliminating labor-intensive debugging of the previous system

Chicago Trading Company achieves 54% cost savings and meets daily SLA for the first time

Recognized as a leading derivatives trading firm, Chicago Trading Company (CTC) provides liquidity to markets around the world, helping drive efficient, stable and healthy markets by participating on both buy and sell sides. CTC’s research platform collects information from thousands of sources, including feeds from every exchange it trades on, historical trading prices and third-party data. But CTC was paying $800,000 a year just to move data from Snowflake to managed Spark for processing and back again. 

To overcome these hurdles, CTC moved its processing off of managed Spark and onto Snowflake, where it had already built its data foundation. With Snowflake and Snowpark, CTC has gained greater visibility and control over its costs while also drastically reducing data processing job failures — an invaluable improvement, given that the jobs are always running against a clock. Thanks to the reduction in costs, CTC now maximizes data to further innovate and increase its market-making capabilities.

Benefits include: 

• 54% cost savings — amounting to millions of dollars annually — by moving from managed Spark to Snowflake

• $800,000 saved annually by eliminating data movement out of Snowflake and back

• First time meeting the daily service-level agreement of having data available at least one hour prior to market open — a milestone it hadn't been in a position to track before Snowflake 

Swire achieves millions in cost savings and accelerates model deployment by weeks 

Swire Coca-Cola, USA is the local bottler for Coca-Cola and other beverage brands in 13 states across the American West, delivering refreshments to 31 million consumers every day. Swire had Snowflake as its single source of truth and a separately managed Spark platform for its AI/ML needs. But managing complex infrastructure diverted data teams from model building, causing delays. Spark clusters needed manual maintenance to avoid waste and took 10-15 minutes to spin up, while the managed Spark platform outside Snowflake raised data governance concerns, impacting data integrity and security.

Snowflake emerged as the ideal one-stop shop for Swire’s AI/ML needs, offering a singular platform that significantly reduced complexity, enhanced ease of use and provided a robust framework for improved data governance. With these improvements, Swire has optimized its planned logistics routes to significantly reduce costs related to fuel, driver expenses and overall cost to serve. The impact on time to market was equally remarkable, with Swire able to develop models on Snowflake notably faster.

Benefits include: 

• Millions of dollars in cost savings by optimizing planned logistics routes

• Faster time to market, resulting in weeks of time savings, by deploying critical AI/ML models faster 

• Lower total cost of ownership from streamlined, automated data management

More migration successes 

These stories are just the beginning of how organizations are moving to Snowflake to drive competitive advantage. 

Download the book “Secrets of Apache Spark to Snowflake Migration Success” to see the five key reasons companies are moving to Snowflake — and how these migrations are helping businesses slash costs, reduce complexity and improve reliability for their daily operations.

LPP leverages data feeds on reported leaked passwords from industry-leading threat-intelligence providers. Snowflake then securely verifies whether the leaked password is still valid for the identified user. Our LPP system validates passwords for all users (human or service) in a privacy-preserving manner. Passwords are only handled in the memory of our automated systems and at no point do they persist at rest in cleartext, nor are they visible to any Snowflake employees.

Once a leaked password is confirmed to still be valid, LPP automatically disables the password for that user. The user will then need to contact their account administrator to get a reset password link which requires them to change their password on next sign-in, which will then be subject to the effective password policies on that account. We strongly recommend that admins turn on multi-factor authentication (MFA) (if the user is not a service user) and network policies immediately. LPP keeps the user and relevant administrators informed, via email, about any actions taken. 

While we continue to believe that MFA is the best protection for user accounts and we will continue to default to MFA for human users, LPP is an additional step toward helping our customers better secure their accounts by default. 

To learn more about how we are making the Snowflake platform more secure and the role of Snowflake Horizon Catalog, watch the BUILD 2024 “What’s New” session on demand.

Since launching the public preview of Hybrid Tables this year, we have seen adoption across industries from customers such as Siemens, Panther, Mutual of Omaha, PowerSchool, MarketWise and Project Lead The Way. These organizations and many more are using Hybrid Tables to simplify their data architectures and governance and security by consolidating transactional and analytical workloads onto Snowflake's single unified data platform.  

Managing application state and metadata

Use Hybrid Tables as the system of record for application configuration, user profiles, workflow state and other metadata that needs to be accessed with high concurrency. Customers such as Siemens and PowerSchool are leveraging Hybrid Tables to track state for a wide variety of use cases.

“Siemens leverages Hybrid Tables in its Data Ingestion Engine to overcome concurrency challenges and improve data quality and consistency for its critical ERP replication process,” says Henrique Dias, Service Manager and Data Architect at Siemens AG. “With Unistore and Hybrid Tables, we can further scale and support our growing Snowflake-based Siemens Data and AI Cloud.”

The low-latency and high-concurrency capabilities of Hybrid Tables also power new customer experiences built directly on Snowflake. For example, PowerSchool, a leading provider of software for K-12 education, needs to deliver secure and efficient solutions that help its customers manage and analyze petabytes of data from disparate systems to maximize educational achievement.

"We are exploring the potential of Hybrid Tables to simplify workflows and meet high-concurrency, low-latency demands of PowerBuddy, our responsibly designed AI assistant," says Shivani Stumpf, Chief Product and Innovation Officer at PowerSchool. "This innovation could allow our customers to engage with millions of parents and students with a transformative user experience.”

Serving data to applications with low latency

Build customized user experiences, dashboards and reports that quickly load precomputed data from Hybrid Tables into applications. With fast point lookups in Hybrid Tables, you no longer vneed to move data to a different database to power low-latency user experiences, which reduces complexity and speeds up development cycles.

Roofstock, a leading investment platform for single-family rentals, is seeing these benefits firsthand. "We are using Hybrid Tables as the backbone for our Data Services use cases,” says Ken Ostner, SVP of Data at Roofstock. “Snowflake’s Hybrid Tables allow us to simplify our data architecture, eliminating the need for data replication and eventual consistency in our environment, ultimately helping us speed up response times for end users and simplifying our analytics pipelines.”

Developing lightweight transactional applications

Open up new possibilities for developing applications on Snowflake with Hybrid Tables’ transactional capabilities. With low-latency point operations, powered by a new row-based storage engine, you can simplify application development by building on Snowflake's cloud data platform. Apps built on Snowflake also gain the advantage of Snowflake’s unified data governance model, which helps maintain security and compliance for all your data. Mutual of Omaha leveraged these capabilities when creating its customer journey application for call center agents, consolidating transactional and analytical processing in Snowflake.

"Snowflake's Hybrid Tables have accelerated our ability to manage marketing campaigns in real time, providing us the agility to process customer data and optimize marketing spend efficiently,” says Lorenzo Ball, Chief Data Officer at Mutual of Omaha. “By integrating transactional and analytical data, we've streamlined operations, eliminated data replication costs and boosted responsiveness to customer interactions. This technology empowers us to act swiftly with the insights we need, all while maintaining the security and scalability that Snowflake delivers." 

What’s new in GA?

As part of making Hybrid Tables generally available, we have incorporated enhancements in the following key areas. 

• Improved performance: Our testing shows a 50% reduction in p99 latencies and price performance for point operations.

• Faster and more cost-efficient bulk loading: We’ve seen up to a 10x improvement in both bulk load speed and cost.

• Larger capacity limits: 1 TB database sizes are the default, with larger sizes available upon request.

• Enterprise features: Additions include continuous backup/restore with Time Travel to easily query older versions of your Hybrid Tables data. 

• Improved monitoring experience: A grouped query history dashboard and additional query profile metrics for Hybrid Tables are also available.

Getting started: Tips and resources

Hybrid Tables are available today in Snowflake databases on AWS. You can start using them immediately — simply specify the HYBRID keyword when creating a new table:

For more information and examples of how to get started, check out these resources:

• Watch the What’s New session from BUILD 2024 and learn more about Hybrid Tables in our documentation.

• See how PowerSchool, MarketWise and Project Lead The Way are using Hybrid Tables to simplify workflow state and meet high-concurrency, low-latency requirements.

• Watch a demo of a low-latency data serving with Hybrid Tables.

• Build your own to-do list application in five minutes.

• Create a fully functional order management app for food trucks.

What's next for Hybrid Tables?

Hybrid Tables are a transformative capability that combine transactional and analytical capabilities in a single database and bring the simplicity and power of Snowflake to operational data workloads. We’re incredibly excited about the new possibilities we see customers discovering. In the future, look for continued enhancements to Hybrid Tables, such as deeper integration with existing Snowflake features, and expansion across Snowflake cloud regions.