Denise Persson, CMO of AI data cloud solutions provider Snowflake, spoke to tipsheet.ai yesterday about the details of the recently announced partnership with agency holding company IPG and its data management unit, Acxiom.
Termed a “strategic initiative,” the companies hope to provide a solution for marketers that addresses consumers in real-time and will involve connecting IPG, Acxiom and their clients to Snowflake’s AI data cloud capabilities as well as the rapidly evolving innovations in AI more broadly.
Elements of the partnership include a human component, too. With the “Joint Solutions Hub for Continued Innovation,” Snowflake, IPG and Acxiom are planning to brainstorm “new AI and omnichannel marketing solutions” and include training on, and consideration of, the latest advertising and marketing technology.
Explore…
- Importance of Snowflake’s partnership with IPG and Acxiom from an AI perspective
- Omnichannel marketing and AI
- AI’s effect on Snowflake’s ad tech sector go-to-market strategy
- Snowflake’s Cortex integration with IPGs Interact platform and where AI fits in
- From Snowflake’s perspective, what should the marketer be thinking five years from now?
- Is there an agentic AI strategy for Snowflake?
Scroll down for the interview which has been lightly edited for clarity.
On the importance of Snowflake’s partnership with IPG and Acxiom from an AI perspective
Ms. Persson: The biggest challenge that marketers are facing right now is data silos and fragmentation -the data lives in so many different applications. And with that, you can never get that 360-customer view and experience you want.
Marketers want one source of truth for their customers. And it’s not only critical to get that view, it’s also essential for AI.
If that data isn’t kept in one place, if it doesn’t have the right inner-governance, if it’s not of the right quality – you can’t do anything with AI either. It’s also important to remember that there is no AI strategy without a data strategy.
So, the marketers and companies that have their “data estate” in order are the ones who can move fast, succeed and are getting the benefits from AI.
Regarding this partnership with IPG and Acxiom, traditionally, marketers would send their data to a data partner and that partner would do the processing, data enrichment and all those various components -many would refer to this as sending something to a “black box” …and marketers really didn’t know what agencies were doing with the data, but they were getting something back.
Now, the Acxiom platform will run natively in the customer’s own data environment. So, if a big brand has different data privacy settings, data governance and so on – that stays intact.
When data moves somewhere, there are many issues including a cost of moving data and quality problems. There’s a real-time problem, too. With the Acxiom platform, it is going to run directly within the customer’s environment, everything is done in real-time and that’s where marketing is 100% moving to – real-time.
You want to be able to respond to customers immediately. There’s a lot that Acxiom is going to be able to do from an AI standpoint, as well, when the platform runs directly in the customer’s environment.
On omnichannel marketing and AI
Ms. Persson: I think we’re definitely in an omnichannel [marketing] world. But we don’t want to be in an omnichannel data situation -that’s a problem.
If all those “omni” channels create new data silos for the marketer, then there’s no way to see what is impacting what.
As marketers, we continuously want to optimize our mix. I have an omnichannel marketing mix, but I want to optimize that and consider more spend in one channel versus another and how things are impacted – like ROAS in the business-to-consumer space.
But, if you don’t have everything in the same view – in one source of omnichannel “truth” and see how it’s performing – you can’t make the right decision about how you’re going to invest.
Where we’re moving to is those brands that have an omnichannel “mono” view today , if you will, will be able to optimize their spend in real-time. In the past, we’ve been close to this, but we still didn’t know until after the campaign is over.
Ultimately, we’re moving towards a future when all of this is done automatically for you. That’s the “Holy Grail” – where you don’t need to sit and do the optimization yourself. We’re going to have this fully AI-based, self-optimizing system, doing all the media, all the content creation, hyper personalization…. and in real-time. That’s exciting.
As a marketer myself, there’s no better time in history to be marketing. Marketing has always been about relevance and timing, and AI is finally helping us with those two things.
On AI’s effect on Snowflake’s ad tech sector go-to-market strategy
(One element of the partnership with IPG and Acxiom is additional partnerships with ad tech firms Mediaocean – and their Prisma platform – and The Trade Desk.)
Ms. Persson: First, for Snowflake, it’s right around our 10-year anniversary in terms of addressing ad tech, our first segment.
This was such a data-intense segment and we worked to solve all the problems that ad tech companies have. At the same time, ad tech helped drive innovation for Snowflake with their advanced data processing needs.
Today, our customers are across industries, but they’re benefiting from how ad tech pushed Snowflake to do the things that they needed Snowflake to do.
Regarding AI, [Snowflake and ad tech] need to solve the AI challenge for customers and integrate innovation for them. We need to make it very easy, too. Everyone wants something turnkey – after all, marketing departments don’t have hundreds of engineers. And yet, brands want to take control of their data for themselves.
There are many reasons for that.
AI is driving a bigger need for data privacy rights. The concern for the big brands is what *could* happen with their data and, of course, they never want to jeopardize the trust of the end customer. AI has been a tailwind for Snowflake in this regard because it drives the need for data governance and data privacy compliance in a whole different way -after all, the marketer wants to use the best-of-breed applications driven by AI.
On Snowflake’s Cortex integration with IPGs Interact platform and where AI fits in
Ms. Persson: Cortex is our framework that allows you to bring any LLMs (Large Language Models) into your data – OpenAI, Anthropic, Mistral and others which each bring different strengths. You can have them run on your own enterprise data.
So, think about ‘talking’ to your data – the dashboards are going away. You will be able to ask a question and get the full answer. For Acxiom, they can use it as well… the AIs, the AI processing. That’s why it’s such a great environment to build on.
From Snowflake’s perspective, what should the marketer be thinking five years from now?
Ms. Persson: The interesting thing is that, of course, we really don’t know what is going to happen. But, what we do know is that you need to have a data strategy, build competence on your team, up-skill your team – you need to make sure you have a culture that fosters experimentation and innovation. You cannot wait and see.
Those companies that are iterating and testing, those are the ones that are winning.
At Snowflake, we have identified people in every single function that essentially are our AI experts, and they’re tasked to go out and test things, experiment, and then on a quarterly basis, they’re presenting to everyone else on what they’re finding.
You need to have a plan.
Is there an agentic AI strategy for Snowflake?
Ms. Persson: Absolutely. In fact, an announcement that we made a few weeks ago was the acquisition of a company called “Crunchy Data” that is in the postgres database space.
You need to have a transactional component of your data platform to be able to build these agentic applications – autonomous, real-time workflow applications that are going to be able to make decisions.
Think of it as a “self-driving” marketing platform in the future and “agentic” is what we are really investing in at the moment from a platform perspective so that companies like Acxiom can build agentic applications for brands.
Snowflake is the infrastructure and we provide Acxiom the ability to build that future where there is a fully automated platform doing all the media planning, media buying, content creation, segmenting, hyper personalization and so on. That is the future -and that’s an agentic application.
I also want to add that we have to think about the consumer here. As a consumers, we have been bombarded by irrelevant content. It’s tiring, wasteful. Think about a future where things are much better. That that’s going to be a bright future, both for consumers and for brands.