What you need to know about data in marketing

Google has changed course — yet again — on its plans for third-party cookies, announcing on April 22, 2025, that it will continue to offer the tracking technology in Chrome, rather than developing a standalone mechanism around their usage, as previously planned. For marketers, the news is mixed, potentially addressing issues around reach in a fragmented media landscape while also lighting a fire under internal plans to safeguard customer data. 

The latest move appears to put the final nail in the coffin for cookie deprecation while possibly teeing up a future in which Google has offloaded Chrome due to regulatory pressures. Many factors were behind the shift, per a company blog post, including feedback from publishers, developers, regulators and the advertising industry about its Privacy Sandbox proposals, the cookie alternative initiative it launched in 2019. A Google executive expanded on the thinking behind the decision when asked for comment at the IAB’s Public Policy & Legal Summit on April 22.

“When you look at the evolution that the whole industry has gone through, and the regulatory environment, all the pieces from when we first announced Privacy Sandbox a number of years ago until today, it’s continuing to evolve and develop,” said Ghita Harris-Newton, senior director for government affairs and public policy at Google. “We’re continuing to work with the industry and think about that evolution and respond to the feedback that we’re getting from the industry.”

The tech giant in July 2024 said it would explore a “new path” around online privacy instead of deprecating third-party cookies in Chrome, halting plans that first began in 2020. Despite countless resources being spent throughout the industry to develop plans for a cookieless future, the company has now decided to keep cookies embedded in Chrome.

The news was announced days after Google lost a second antitrust lawsuit, with a federal judge ruling that the company wields an illegal monopoly over key aspects of online advertising technology. The company in August 2024 was found to have an illegal monopoly over search, a case that could see the company forced to sell off its popular Chrome web browser. The regulatory pressure could be a factor in why Google is again rethinking its third-party cookie plans.

“This news isn’t surprising. They likely don’t want to do anything that would potentially hurt their ads business if they are ultimately forced to sell Chrome,” said Amelia Waddington, global chief product officer at search intelligence firm Captify, in emailed comments. “Sandbox worked because it was based on them owning Chrome and having complete control of the supply chain. If they don’t have Chrome, they will need cookies for targeting, just like everyone else.”

The industry responds

While Google will maintain its current approach to offering third-party cookies, it isn’t completely abandoning privacy-preserving solutions. The company will continue to enhance tracking protections in Chrome’s Incognito mode and plans to launch a new IP Protection feature in Q3 2025, per the blog post. It will also gather industry feedback about what role the Privacy Sandbox will play, sharing an “updated roadmap” and plans for future investment in the coming months. 

As with previous developments in the long, often hair-raising saga around cookie deprecation, opinions on how Google’s latest move will affect the ad industry and how marketers should respond are mixed. While some, like IAB CEO David Cohen and IAB Tech Lab CEO Anthony Katsur, hailed the move as a win for interoperability or a confirmation of where the ad industry is already headed, others decried it as undermining years of industry progress around privacy.

“At a time when consumers are more aware than ever of how they’re tracked online, doubling down on cookies sends the wrong message. The internet doesn’t have a targeting problem; it has a trust problem,” said Adam Schenkel, executive vice president of global platform strategy and operations at contextual targeting firm GumGum, in emailed comments. “Brands should take this as a wake-up call to diversify their ad strategies now, before audience trust erodes further.”

Other executives noted that marketers have already moved on from third-party cookies amid industry-wide attention on first-party data and channels like connected TV and mobile that are already cookieless. 

“While we expect to see marketers continue to rely on third-party cookies indefinitely for reach, we also expect an increased emphasis on tools that can incorporate the third-party cookie alongside newer destinations and experiences, and provide interoperability and a unified view across all,” said Travis Clinger, chief connectivity and ecosystem officer at LiveRamp, in emailed comments.

More scrutiny

While the questions over the future of third-party cookies in Chrome come to a close, Google is far from out of the woods in its antitrust battles. The course changes and regulatory pressure could encourage the ad industry to continue pushing the largest player in digital advertising for more transparency and competition in other areas.

“Last week’s second major legal judgment against Google for monopolistic practices is expected to ‘reshape’ digital advertising. It also gives marketers the chance to let Google know that the end of opacity in their video advertising practices should fall next,” said Sean Cunningham, president and CEO of the VAB, in emailed comments. 

“While marketers know that the depth and breadth of Google’s non-transparency in video advertising is not (yet) a crime, now is the time to say ‘it ought to be’ and demand and end to the industry’s double standards in transparency, that dynamic certainly needs ‘re-shaping’ as well,” the executive said.

More than five years after Google first announced plans to deprecate third-party cookies, the reshaping of the digital advertising landscape is well underway. The ad industry has been hard at work, adopting clean rooms for data collaboration, developing IDs for better targeting and refining marketing mix models. For the forward-looking players involved in these efforts, the latest news is a “moot point,” said Mari Docter, senior vice president of data strategy and innovation at indie media buying and planning agency Novus.

“The industry is moving towards a more privacy-compliant way of using data, and as all of the fragmented state by state privacy laws come to fruition, the cookie will just become a meaningless piece of technology moving forward,” Docter said in emailed comments. “We are not going to go back in time.”

Only 30% of agencies, brands and publishers have fully integrated artificial intelligence (AI) across their media campaign life cycles, per the IAB’s State of Data 2025 report. While half of those who haven’t yet fully integrated AI across their media campaigns expect to do so by 2026, most do not have a strategic road map. Half of brands are also concerned about a lack of transparency into how agency and publishing partners use AI on their behalf. 

Additionally, one of the largest obstacles to adoption are data concerns, with nearly two-thirds of those queried citing the quality of the data being used and produced, the protection of that data and fragmentation among disparate AI tools as top barriers. The report from the IAB was released in March 2025.

For all of the discussion about AI and the spending on new tools, practices and methods to harness it, it has yet to reach widespread use. However, as the tech continues to rapidly evolve, those who are slow to get on board are at risk of being left behind, according to Angelina Eng, vice president of measurement, addressability and data center for the IAB.

“The industry isn’t taking full advantage of what AI can do now. It can build media plans, generate audience segments, select media partners, forecast performance, and even use synthetic or ‘fake’ data to enhance marketing mix modeling and sales attribution,” Eng said in a statement. “This will transform how media campaigns are conceived, executed, and managed from start to finish.” 

However, of those using AI, the majority are very satisfied. Seventy percent of agencies and publishers, who are ahead of brands in terms of adoption, say the tools meet or exceed their expectations for driving time, cost savings and resource goals.

The report suggests there is still a long way to go for widespread and enthusiastic AI adoption. A majority of agencies, brands as publishers (as many as 90%) are using general-purpose tools such as ChatGPT, Perplexity and Meta AI — or functions within the platforms they’re already using, such as The Trade Desk’s Koa, Google Smart Bidding and Merkle’s GenCX — for their AI use. While these tools can provide an introduction and easy access to AI use, they don’t always have the robust functionality that will lead to a widespread business advantage.

Further, only about half of the agencies, brands and publishers responding are using or planning to use any of the 18 solutions proposed by the surveyors — such as long-term strategic roadmaps, clear use cases or defined KPIs — to overcome the AI adoption challenges for their media campaigns. Additionally, only 20% are putting in place AI boards or contract clauses, which will leave them open to biased outputs, data breaches and compliance failures. 

Ultimately, those who don’t have plans in place will be susceptible to falling behind their industry competitors, losing consumer trust and facing legal consequences.

“The study suggests that agencies, brands, and publishers need to take a phased approach to AI adoption, ensuring that data inputs and outputs are secure and of high quality,” Eng said. “They also need to train teams on best practices, collaborate with the industry to develop standards and prioritize key use cases to establish a strong foundation for full-scale adoption.”

The IAB partnered with BWG Global and Transparent Partners for its report, which surveyed over 500 subject matter experts at agencies, brands and publishers.

Sponsored content

By Amperity

The triple convergence

We’re in the eye of a storm where three powerful forces are colliding: the explosion of customer data, the rapid advancement of artificial intelligence (AI) and intensifying economic and geopolitical uncertainty. 

1. Economic uncertainty breeds changing consumer behaviors and budget constraints
2. The explosion of customer data is making it hard to find the right signal
3. Rapid advancement of AI is allowing brands to more efficiently reach customers

This convergence is reshaping the way businesses must operate — and creating a new imperative: a strong customer data foundation built to help brands target and retain customers through economic headwinds and ready to take advantage of AI opportunities. 

Economic pressure demands operational agility

Economic pressure is forcing brands across industries to re-evaluate how they operate, prioritize efficiency and maximize the value of every customer interaction. The changes businesses are making during these uncertain times are not temporary — it’s a permanent shift. 

• Retail: Margins are being pinched with rising costs, and customers are getting more budget-conscious. According to Reuters, consumer sentiment is at a three-year low due to recent uncertainty.
• Travel and Hospitality: With price sensitivity and unpredictable booking behavior, personalized offers that work are more critical than ever. A Skift survey finds international travelers are less likely to visit the US.
• Automotive: Supply chain unpredictability and evolving buyer preferences mean the margin for error in customer targeting is razor thin. According to S&P Global, “auto tariffs could impact as much as 45% of light-duty vehicles sold in the US.”

In each case, companies must do more with less. AI offers a path forward — but only if their data foundation is in order. 

The customer data crisis

Brands are collecting five times more customer data than they did just four years ago, according to IDC. Every click, purchase and customer service interaction creates a digital breadcrumb — yet most organizations still lack a coherent customer view. Leveraging an inaccurate or incomplete customer data foundation means:

1. Customer insights that are used to drive overall strategy are misleading your business.
2. Customer experiences that are meant to engage and grow the value of your customers are not relevant, often having the opposite effect.
3. AI investments meant to automate and accelerate key initiatives are making decisions on misleading data.

Why do brands struggle to create a unified view of their customers? Because the majority of innovation has been happening in analytics and personalization, while the hardest problem of data unification has barely evolved. Legacy systems, built around siloed channels and static rules, simply can’t keep up with today’s data complexity and scale. And as the adage goes, garbage in, garbage out. This results in suboptimal revenue growth and the inability to take advantage of the next generation of AI.

AI has arrived. Are you ready?

AI capabilities are advancing at breakneck speed. AI agents have the potential to deliver one-to-one personalization at scale that has eluded most brands. Machine learning models can spot patterns across billions of data points. Large language models understand context and intent to serve customers the best possible experience. 

In the face of tight budgets and rising pressure, AI isn’t a luxury — it’s a necessity. C-suite executives have drawn the line that the time for AI adoption is now. However, according to MIT Technology Review Insights, 78% of global companies are not “very ready” to deploy natural language AI tools like LLMs and agents. The biggest barrier? Data readiness.

Why customer data is the bottleneck

While many teams rush to apply AI to personalization or insights, the real bottleneck to seeing the expected returns from these investments is further upstream: messy, mismatched customer data.

The explosion of customer data comes from a proliferation of channels and applications, each defining customer identity and capturing data differently, which will only get more complex with AI. And customer data has always been complex — people (and their data) change — they move, get married, sign up for multiple accounts with slightly different names and new emails. Add to that the massive volume, human error typos and siloed systems and you get the reality most teams face: disjointed, unreliable data that undermines even the most advanced AI initiatives.

Build an AI-ready data foundation

Enterprises serious about AI transformation must begin with an AI-ready data foundation:

• Accurate, unified customer profiles with transparent matching methodologies
• Flexible data models that can both incorporate new data sources quickly and shape data as the foundation for new AI initiatives and activation channels
• Governance and access controls that ensure responsible AI use

Without this foundation, AI becomes another disappointing investment. Because even the smartest model is only as good as the data it’s trained on.

AI’s most urgent job: identity and data management

What organizations need first isn’t customer-facing AI Agents — it’s Agentic AI for the tedious, complex work of customer data stitching. There is an opportunity for AI agents to lighten the burden of complex code for Identity Resolution on data engineering teams, allowing them to focus on strategic decisions around model tuning and data QA.

Without clean, connected customer data, every downstream AI use case — no matter how sophisticated — will be built on shaky ground. Identity resolution isn’t just an operational step, it’s the linchpin of AI-driven transformation.

The path forward

The future of customer data is not just about AI — it’s about how prepared your organization is to use AI where it matters most. It starts with a strategy to maximize the value of customer data, a modern AI-powered platform to accelerate data readiness and strong data governance.

The question is no longer whether AI will transform customer data foundations — it’s whether your organization will lead or follow in this transformation. The best customer data wins.

To learn how Amperity is tackling data readiness with our new Identity Resolution Agent, check out the demo.

Amazon on Jan. 27, 2025, announced the latest addition to its ad offerings, Brand+, which optimizes connected TV (CTV) and online video ads with the help of artificial intelligence (AI). The solution brings together trillions of shopping, browsing and streaming signals to help brands reach potential customers across Amazon channels like Prime Video and Twitch as well as video publishers including Buzzfeed and Dotdash Meredith.

In a hypothetical use case for Brand+ offered by Amazon, a travel company would upload first-party data to Amazon DSP and the new capability would use machine learning to identify customers who’ve searched for travel gear, bought travel guides and streamed travel shows, delivering ads to these high-potential audiences. 

Some beta testers of Brand+ saw more than a 10% increase in sales and over a 70% increase in website traffic, adding Brand+ to the list of AI-powered tools at advertisers’ disposal, including Google’s Performance Max and Meta’s Advantage+. But more so, Brand+ demonstrates how the convergence of retail media networks (RMN) and connected TV can satisfy marketer imperatives.

“As marketers increasingly focus on precision and ROI, retail media CTV is set to redefine the advertising landscape,” said Dave Peterson, general manager and global head of retail media at Epsilon, in emailed comments. “Prioritizing impactful, personalized campaigns over campaigns that have broad reach enables brands to forge deeper, more meaningful connections with consumers.”

By 2028, CTV ad spending is forecast to surpass traditional TV ad spend and omnichannel retail media will account for nearly a quarter of all U.S. media ad spend, per eMarketer data. The continued growth in and dominance of ad spend by two symbiotic, synergistic channels could finally deliver on digital’s promise of full-funnel, data-driven marketing.

“If you look at what brands and especially Fortune 500s are trying to achieve, it’s speaking to specific people — based on first-party data from the retail media networks — in every touchpoint available for them,” said Oz Etzioni, CEO of advertising personalization platform Clinch. “Before CTV, it was pretty limited... CTV really opened up the full spectrum of how you communicate with people and how you start to take them down the funnel.”

Better data, better partnerships

The last few years have seen an increase in partnerships between retailers and media companies — NBCUniversal with Instacart and Walmart, Disney with Walmart and Kroger, Roku with Instacart, and so on — as both sides look to take advantage of the RMN-CTV convergence. For retailers, CTV opens their media offerings to off-site ad inventories.

“Authenticated data from CTV is hugely valuable, and comparatively, you’re actually able to do better identity matching with that authenticated data, whereas you’re really not able to do so much with authenticated data from social,” Jeffrey Bustos, the former vice president of measurement addressability data at the IAB who recently joined Dentsu’s Merkle as senior vice president of retail media analytics. “We’ll see that a lot more, especially as we continue to lose data signals.”

The authenticated data allows advertisers to reach audiences without the use of only their own deterministic data, expanding the reach from just the consumers who shop at a specific retailer to the millions of consumers who watch YouTube or Netflix, for example.

“As a retailer, I can use [CTV] data to create a lookalike audience, to then target a larger subset of people than just my [audience], and then I can see within that larger subset how many other people went to my store afterwards and bought something, whether it’s online or in store. So that’s the value,” Bustos said.

When it’s time to measure outcomes, the combination of CTV and RMN makes it easier and quicker for advertisers to see how their campaigns performed. The type of work that clean room provider Habu does with Kroger and Disney around audience activation is what advertisers should expect in the future, especially as players like Circana and Dunnhumby begin to offer cross-retail attribution, Bustos explained. But there’s still work to be done to connect the pipes and standardize data.

“We’re pushing on that standardization with the IAB across all different retail media networks as they move up the funnel to be able to lean into conversion metrics and do it in a way that is comparable so that we can look at our media holistically, and determine the best places to spend and trust those results,” said Jill Cruz, executive vice president of commerce strategy at Publicis Commerce.

Personalized creative at scale

CTV has long been bandied about as a channel that can meet marketer needs around personalization at scale. As ad tech in the CTV space gets more sophisticated, especially alongside the rise of AI, advertisers can deliver on that promise with buys powered by retail media networks. A brand using Kroger Precision Marketing, for example, could buy the retailer’s audiences across Roku, Disney, YouTube and beyond — scaling the same way programmatic buys are.

“Within advertising, scale is everything and automation is everything. A retailer or brand using retail media data can now access multiple CTV partners using the same creative so they’re able to scale a CTV buy,” said Bustos.

Apart from versioning and personalization, CTV is likely to see even more creative experimentation as advertisers get more comfortable with the space. Instead of repurposing 15- and 30-second ads, there is an opportunity for agencies to make a splash, Cruz explained.

“Retail media creative has been a place where sometimes creatives go to die,” the executive said. “CTV can change that for retail media. Creative can be a little more custom and speak to that audience in a way that’s more driving into the basket versus just equity building. That’s an area of opportunity in 2025.”

Along with opening the ad experience to creative that is more engaging and personalized, the combination of CTV and retail media gives advertisers more abilities to connect with consumers. 

“You’re looking at different ad units, different ad experiences involving second screen, involving data that comes from the outside, like location, weather and sports,” said Clinch’s Etzioni. “I see a huge opportunity, especially for non-pure-performance players.”

The shoppable future

The growth of retail media and surge in shoppable formats could make the need for measurement standards more pronounced, according to Melanie Babcock, vice president of Orange Apron Media and monetization at The Home Depot.

“We were excited to see media partners elevate the importance of RMNs as one of their key customer segments,” Babcock wrote in emailed comments. “We will continue to see use cases on how RMNs can apply AI to operations, creative development and ad serving.” 

Shoppable TV has been a media player and marketer talking points for years, to the point where “buying Jennifer Aniston’s sweater” has become a cliche in ad world circles. But the rise of AI and other machine learning is closer to making the dream of shoppable ads a reality. For example, video platform Kerv.ai uses AI-powered video analysis to identify and match objects in content with contextually relevant advertising experiences. The company has partnerships with NBCUniversal, Warner Bros. Discovery, Paramount and Disney.

“Leveraging the retail media world is a big partnership program for us,” said Gary Mittman, CEO of Kerv.ai. “Why would [RMNs] not want to provide a value of relevance and engagement to their presentation? That’s where we fit like a glove into that universe.”

Still, the rise of better shoppable ad formats that utilize CTV and RMNs doesn’t mean that they are right for every marketer. Very few consumers are ready to buy a car via their remote, and many consumers have tired of the QR code renaissance.

“Just because the technology exists doesn’t mean we should do it,” Bustos explained. “We have to really think about consumers being really lazy.”

The wild west

If consumers think they have it bad with the proliferation of CTV and streaming platforms, they have nothing on advertisers hoping to make sense of the retail media network space that, by some estimates, now has more than 200 players. While most RMN ad spend is with a few major players, the remainder will have to find a way to offer a competitive advantage to advertisers or find a new way forward.

“You’re definitely likely to see aggregation,” said Kantar senior retail thought leader Barry Thomas. “You just can’t get these dollars from big brands with a B-player media network, so there’s lots of aggregation happening… for the majority, that’s your future.”

While some RMNs have signed up with the likes of Instacart or Criteo to address their market position, that strategy has often left these platforms in a “no man’s land,” unclaimed by either the brand or the agency, Cruz explained. As is often the case, not every business can be Amazon.

“The innovation is going to come from identifying other B2B opportunities, such as selling supply chain capabilities, licensing proprietary tech or providing some other form of data-driven insights,” Cruz said.

The fragmentation of retail, retail media and CTV could help smaller players across the landscape find new revenue streams and partnerships, in a tide-raising-all-ships manner. Advertisers might be able to take advantage of the fragmentation in ways they couldn’t if the channels were completely separate.

“It could be a win for retailers saying, ‘We want to drive younger traffic into the store to purchase our product here,’ as well as CTV being able to utilize those audience to get bigger reach,” Cruz said.

Kellanova had a problem with Special K in the U.K. in 2024. The vitamin- and protein-rich cereal had seen volumes decline and stagnating market penetration. As its premium price went up, indexed at 145% higher compared to the rest of the category, other metrics went down. While awareness was not a problem for the 70-year-old brand, consideration rates began to mirror those of private label brands.

To address Special K’s flagging interest, Kellanova  — a group that markets snacks including Pringles and Cheez-It, along with international cereal brands following the Kellogg Company’s 2023 split into two companies — turned to LiveRamp for a clean room. The technology allows marketers to access and collaborate on various data sets while protecting privacy and has regained industry attention as the need for more customer insights creates friction with mounting data-privacy concerns.

“Clean rooms [for] improving our targeting is one of the big areas we identified as having a measurable impact in our work and our connection with the consumer,” said Louise Cotterill, Kellanova’s global senior director of insights and intelligence.

The traditional marketing approach for Special K had been demographic-based and used one set of marketing creative. But by unifying postcode-level Circana data with 20 million addressable records from Experian, Kellanova was able to better understand household compositions and interests.

“How do we segment and identify [consumers] and go beyond demographics into their purchase behavior, historical behaviors and competitor buying [behaviors]? Are there better predictors of who’s going to buy the cereal?” said Cotterill.

The revamped approach identified two distinct groups of Special K consumers: loyal buyers that the brand needed to encourage to buy one more time to grow the brand, and lovers of the brand who only bought discounted Special K and needed an extra push to purchase the product at full price. Kellanova tailored existing creative to match these cohorts, crafting stronger call-to-action messages for both groups, and deployed it across social channels including Meta and YouTube. 

Brand lift and behavior change

Traditionally, CPG marketers relied on inference to determine which campaigns drove which purchases. That changes when using a data clean room.

“With the clean room, we can do a partnership with a retailer, and we can see exactly who purchased,” Cotterill said. “We’re able to do as close as we can get to closed-loop [measurement], and see which groups are buying. That’s why we’re so confident in this approach.”

Using the tactic, Special K saw a 9% sales lift with price-conscious consumers who hadn’t purchased the brand in 2024 but were brought back at a higher price point. Plus, these consumers tended to buy private label, demonstrating that marketing — not just price reduction — could bring them back to the brand. The marketer saw even stronger results with loyalists, who generated a 36% sales lift by buying more of the brand than usual at a premium price.

In addition, Kellanova ran a consideration impact study to see how long-term consumer behaviors changed. The demographic-based targeting showed no bump in consideration, but the use of the clean rooms showed a lift in consideration three-times higher than the industry average. All in all, the new approach took about three months of work.

“It was a really amazing opportunity to really put all of the pieces in terms of the strategy, the advanced analytics, the really advanced measurement into play, and see how it works for our brand,” Cotterill said.

In-house and beyond

The success of the Special K campaign demonstrates how Kellanova can use LiveRamp’s clean room technology to improve targeting and drive business results. Relying on a mix of in-house data scientists, external tech and data partners is helping to speed up the campaign process for the packaged foods giant. The first project under this method took months to do while similar initiatives now take three or four weeks. 

For the insights to drive outcomes, Kellanova relies on agency partners that know how to unlock data and apply it to decisions around creative and media. In the Special K example, the creative team needed to figure out how to change the call to action in the existing campaign, while the media team needed to change the buy from a platform well-suited for the demographic target (TikTok) to one that favored the clean room data (Pinterest).

“We do the analysis, the business intelligence work and the audience building all in-house, but then we work really closely with the creative and the media partners,” Cotterill said. “It’s very much a partnership, but it’s been amazing having the data scientists in-house, because you just learn a lot more… The closer they are to the data, the closer they are to understanding how the future marketing is going to work.”

Kellanova is applying the clean room strategy to other brands in its portfolio and into different markets. But the work does not produce a one-size-fits-all model, especially as regions around the globe have varying levels of data available, not to mention privacy regulations. Paradoxically, the more rigorous the data rules, the more applicable the work is elsewhere.

“Can we do it in one of the more restrictive, still very important markets, and is that going to work? If so, then I know that I’ll be able to take that to … any other other key markets outside of that,” Cotterill explained. “We intentionally choose more challenging markets so that we know we can scale it.”

WPP on April 3, 2025, acquired data collaboration platform InfoSum, enhancing the network’s AI-powered data intelligence and consumer identity capabilities, the company said in a news release. Terms of the deal were not disclosed.

InfoSum will become a part of GroupM, the holding company’s media investment group, and will be integrated into its proprietary WPP Open platform. Lauren Wetzel will remain CEO of InfoSum and will also take on the additional role of chief solutions officer within GroupM. The deal reunites GroupM CEO Brian Lesser, who joined GroupM as global CEO in 2024, with a company he once led. Lesser was CEO of InfoSum from 2020 to 2024. 

WPP is touting its latest acquisition as a way for clients to move away from identity-based solutions “that depend on decades-old, deteriorating databases weakened by cookie deprecation, platform fragmentation and splintering audience match rates,” per the press release.

With InfoSum’s clean-room technology in-house within WPP, clients will be able to combine their first-party data with data signals from billions of touchpoints to generate new marketing intelligence that can build, train and deploy custom AI models for quickly generating insights and optimizing campaigns. InfoSum’s network includes access to media channels such as Channel 4, DirecTV, Netflix and Samsung Ads, as well as global retail networks and data partners including TransUnion, Experian and Dynata. 

“Directly integrating InfoSum’s global data network and technology infrastructure will allow our clients to create even more value from their first-party data and enable us to train client AI models against the most data, from the most places, at unprecedented scale and speed,” Lesser said in the release. “Our approach recognizes the importance of identity data to today’s marketing strategies while allowing us to take advantage of the limitless opportunities for growth we can create by moving beyond them.”

WPP’s deal comes on the heels of Publicis’ acquisition of identity solutions firm Lotame. Lotame will be combined with Publicis’ existing data-driven marketing unit Epsilon, leading the holding company to reach nearly 4 billion unique profiles globally, or about 91% of internet-connected consumers, according to the company. 

Grooming brand Dollar Shave Club, subscription-based DTC food brand Daily Harvest and business-wear brand Ministry of Supply are among those using Klaviyo’s new B2C CRM platform, which launched more broadly on Feb. 20, 2025. The offering hopes to meet the needs of consumer brands who are looking to cut across silos to drive personalization.

For Dollar Shave Club, a more integrated approach to data-driven marketing comes as the company — which in January 2025 made its first major marketing move after being acquired by Nexus Capital Management from Unilever — prepares for a market that is vastly different from the one it faced at its 2011 founding.

“The customer journey and the value prop was pretty simple [in 2011]. Now it’s much more complex,” said Bill Hudak, vice president of DTC acquisition and retention at the brand. “You have to play the customer centricity and personalization game to win in 2025, and Klaviyo’s platform enables us to do it.”

Klaviyo bills its B2C CRM as a platform built for consumer brands — not business-to-business firms. The platform includes marketing, analytics and customer service functions, all powered by the Klaviyo Data Platform (KDP).

Klaviyo in 2024 saw 34% year-over-year growth, with CEO Andrew Bialecki attributing the company’s strong performance to its ability to bring together data, AI and analytics. For a martech company founded in 2012 that historically focused on email and SMS marketing, the launch of B2C CRM addresses a gap in the consumer-brand marketing landscape, said CMO Jamie Domenici.

“[Brands use] a very channel-first marketing strategy, and we’ve noticed those are really converging and becoming more of a smarter system of interaction with your customers,” the executive said. “B2C companies are talking to their customers across the entire lifecycle, and they need a better way to communicate with their customers.”

Connecting the dots

Klaviyo B2C CRM brings together a new analytics product that is powered by artificial intelligence (AI); expanded tools that help brands run, personalize and automate campaigns; and its Customer Hub, a shopper experiences product that seeks to bring together marketing and customer service. The CRM speaks to the many needs marketers face as consumers expect fast, seamless and personalized experiences.

“For brands to meet these demands, they need to engage innovative companies that help unify fragmented tech stacks in order to deliver cohesive, end-to-end experiences that bridge the gap between marketing, commerce, and customer service,” said Roger Beharry Lall, research director at IDC, in a statement. “Connecting these dots will enable brands to forge deeper customer relationships, drive loyalty, and boost revenue growth.”

In this case, the dots that need connecting are the millions of pieces of customer data that brands accumulate. The CRM offering includes more than 350 pre-built integrations that helps brands connect customer, purchase and behavioral data without complex set ups or data silos, and is built on top of KDP, the company’s data platform that stores customer event data, for free and forever. 

“You’re only as good as what you do with the data,” Domenici said. “[Marketers] have to manage like 17 tools and have two people, who don’t have PhDs in data. Our mission is to build out the one-stop shop that makes it easier to drive personalization.”

Marketing — or project management?

Dollar Shave Club, the erstwhile direct-to-consumer disruptor, has seen impressive results after using Klaviyo’s B2C CRM for just a few months, including reducing cancellations by a third and increasing average order value for account holders by 12%. The ability of the platform to consolidate the company’s tech stack made it a “no brainer,” explained Hudak. 

When the executive joined Dollar Shave Club in September 2024, he was impressed with what the company was doing with personalization and segmentation, but was concerned that the marketing team was spending too much time and too many resources bringing everything together.

“Are we doing marketing or project management over here?” he remembers asking.

Dollar Shave Club had already been using Klaviyo’s services. Turning on the CRM functionality was as simple as pushing a button, and a crucial recency, frequency and monetary (RFM) analysis was available “out of the box.”  Since then, the company has used the platform’s data insights to improve its order notification and involuntary churn campaigns; a reactivation campaign using a built-in Meta audience integration generated a 2x return-on-ad-spend. 

The one-stop-shop nature of Klaviyo’s B2C CRM compares favorably to most marketing stacks, Hudak explained, adding that Dollar Shave Club is just “scratching the surface” of making the most of the new platform.

Nearly two-thirds of marketers increased their investments in retail media in 2024, even with lackluster performance and measurement concerns, according to a report from Forrester. With the vast majority of retail media networks launching in the past five years, the space has become crowded with competition. Additionally, the retail media ecosystem has largely failed to deliver on promises surrounding data usability, leaving many marketers frustrated. 

“There’s a divergence between retail media, sponsored content and its reality,” said Nikhil Lai, senior analyst at Forrester. “There’s a gap, and to close the gap retail networks have to act more like media publishers than retailers.”

“The State of Retail Media, 2025” evaluates previous reports and incorporates supplemental information from numerous organizations to predict how the retail media industry will look in 2025. Additionally, Forrester looked at Google trends, SEC filings and other data to update the report for 2025. 

It’s Amazon’s ocean

Amazon Ads’ business is so large that the actual size and growth of the retail media market has been misconstrued. The e-commerce giant’s media network is estimated to be larger than all other U.S.-based networks combined, and removing it from the equation dramatically reduces the size of the overall retail media market, according to Forrester. 

In 2023, Amazon Ads saw a 24% year-over-year growth in business, hitting $47 billion. Walmart was a distant second, only capturing $3.4 billion. This reflects 28% YoY growth for the big box retailer, and includes trade funding, which often shrinks as retail media network spend grows. 

Not only is Amazon the largest retail media network in the U.S., it also has significant technological and data advantages. The e-commerce giant holds the rights to many popular entertainment franchises, such as James Bond and the television rights to Lord of The Rings, allowing for premium advertising placement. It’s a value proposition not many organizations can compete with. Additionally, Amazon Marketplace allows third-parties to sell products, allowing for a vast array of products and demographics and a built-in ad revenue segment. 

“They have a huge marketplace of third party sellers who dedicate all their ad spend to the marketplace that they sell on. So Amazon is unique in so many ways, and it’s misleading for other retailers to think that they can unlock the kind of high market income Amazon does,” said Lai. “That’s where the distortion comes from.”

While the problems facing the broader retail media network segment run far deeper than Amazon, the e-commerce giant is certainly exacerbating them. For example, Amazon offers marketers an easy-to-use platform with tools similar to Google Ads, which many in the industry are already familiar with. However, the majority of other retail media networks remain manual. This, combined with the lower impressions delivered by these other networks, often doesn’t make the effort worth it for advertisers. Additionally, in-store measurements are treated like an afterthought for many retail media networks, according to the report. Some retailers who have invested in smart-shopping tools such as digital displays haven’t rolled them out nationally due to concerns surrounding return-on-investment. 

Changing perspectives

If retail media networks want to be successful, they need to stop thinking like retailers and more like publishers, according to the report. Granting marketers more flexibility, increasing automation and industry standardization are crucial if networks want to continue on a growth trajectory. 

While access to first-party data is one of the things which has made retail media networks appealing, as the space continues to grow, data alone isn’t going to be enough for marketers. If retailers want their networks to remain appealing, they need to do more to set themselves apart from the crowd. Ease of use, increased on-site advertisements and automation are the keys to capturing industry attention, according to the report.

“There’s a couple hundred retail marketing networks out there. And you may have good data, but your data may be duplicative to what someone else has…so data is insufficient,” said Lai.