How AI can solve the pain of growing retail media

Retail media has emerged as one of the fastest growing advertising channels in recent years, but new data suggests this growth is beginning to slow down. According to IAB’s 2025 data report, retail media is projected to increase by 15.6% in 2025, representing a significant slowdown from the 25.1% growth seen in 2024.

This slowdown comes at a critical moment for the retail media ecosystem. Despite continuing to outperform overall advertising growth (projected at 7.3% in 2025), retail media networks face challenges that threaten momentum. As brands demand greater standardization, more refined measurement capabilities and better cross-platform integration, artificial intelligence is perhaps emerging as the most promising solution to these growing pains.

However, implementation is not always easy.

The Perfect Storm: Why is Retail Media Slowing Growth?

IAB data confirms that many industry insiders are observing. Retail media is entering a more mature stage where growth is becoming more difficult to maintain. This is no surprise given the explosive expansion of channels over the past five years. Hundreds of retail media networks are competing for advertisers’ budgets.

Several factors contribute to this slowdown.

Ecosystem fragmentation: Brands with over 70 retail media networks in North America alone struggle to manage relationships across multiple platforms, each with their own interfaces, metrics and workflows. Cost rise: As competition for premium positions increased, the costs of retail ad units (CPC and CPM) have risen significantly, causing brands to question the ROI. This cost pressure is becoming more and more explicit, as evidenced by my recent report on Walmart’s approach. Retail giants are pushing their brands to increase retail media spending by 25% year-on-year, as part of their joint business plan, despite many brands reporting stagnant sales growth from these investments. Measurement Challenge: There is no standardized measurement across the network, making it difficult for advertisers to compare performance and justify their investment. Integration Complexity: Brands want seamless integration between on-site and off-site inventory, but many networks still operate these as separate channels.

As a result, advertisers still value retail media purchase points and their proximity to closed-loop measurement capabilities, but many are becoming more selective about where retail media dollars are allocated.

How AI is dealing with the pain of growing retail media

IAB’s Data Status 2025 Report provides potential paths to advance through artificial intelligence. According to the report, 80% of buyers have already used or investigated the generation AI tools to plan and activate the media, while agents are leading adoption by 83% compared to brands, compared to 71%.

Especially in retail media, AI applications are deployed throughout the campaign lifecycle to address key issues.

1. Smarter planning and audience development

The IAB identifies AI-driven scenario planning as an important opportunity for brands and agencies. Using AI, buyers can simulate different budget allocations across retail media networks and predict results before launch. This helps optimize spending across the fragmented ecosystem by identifying which networks provide optimal performance for a particular campaign goal.

Newstream Media Director Alex Arnott has consulted with retailers on media offerings and says the industry is in the early stages of understanding how it can be equipped to automate and enhance AI. He calls buy-side ad buying platforms like Skai and Xnurta. It uses AI to derive omnichannel insights and automates the development of omnichannel media plans.

2. Optimizing your automatic campaigns

For Activation, IAB recommends AI-driven automation that dynamically coordinates bidding strategies, pacing, and creative rotation across multiple retail media networks. The IAB Data Status 2025 report highlights how AI can enable “AI-driven campaign orchestration and content optimization.” The brand emphasizes “automatically adjusting tactics” to monitor real-time performance and improve results while “utilizing and launching campaigns across paid, owned and acquired channels.”

Vince Crimaldi, Capgemini’s retail market unit leader, says AI algorithms can analyze real-time sales data within a given network and automatically transfer budgets to higher transformed products or ad placements.

3. AI-led creative

One of the earliest AI adoption of markets was the development of written, visual and video creatives for advertising. Alex Arnott of New Stream Media says that many retail media networks use Genai to provide brands with self-service, creative products. This allows brands to quickly and efficiently develop landing pages powered by their own creative assets and AI (rather than adding creative production fees for RMN to develop). “The RMN advertising platform is jumping on the bandwagon as it incorporates ready-to-use Genai creative capabilities into its platform to encourage long-term tail brand investments,” says Arnott.

This feature is especially powerful when you are integrating with the automated campaign optimization strategy mentioned above and creating a more responsive and intelligent advertising ecosystem.

Reality Check: Why Is AI Hard to Implement in Retail Media?

While AI holds promises, retail media faces clear hurdles that make implementation more difficult than other digital advertising channels.

The challenges of technical fragmentation: Unlike other digital advertising channels, retail media suffer from basic infrastructure issues. Each retailer maintains a unique data ecosystem with a variety of taxonomy and metrics, creating a major barrier to the implementation of cross-network AI. While major players like Amazon offer robust APIs, many small networks lack the technical infrastructure needed for comprehensive AI-driven optimization. This fragmentation makes it extremely difficult to develop AI models that can provide meaningful insights across multiple retail media platforms.

Clean and Organized Data Sets: Neil Sheridan, an RMN expert who played a leadership role in Macy’s and Berg’s RMNS, says access to first-party data can enhance personalized advertising and targeting, but the effectiveness of AI depends on clean and organized data sets.

Retailers often struggle with data silos, inconsistent quality and integration issues, Sheridan says it complicates accurate analysis along with concerns about consumer privacy. “Companies also need to invest in robust technology infrastructure and rethink how they measure campaign effectiveness,” Sheridan says.

Data privacy and competitive concerns: Retailers face serious challenges. The need for data privacy balance and comprehensive insights in AI. The isolated nature of the data environment of retail media creates a paradox that limits the possibilities of AI by protecting shopper information. As Sheridan points out, “Stakeholder collaboration is essential for effective data sharing while maintaining consumer trust.”

What is realistically possible at the moment

Several priorities emerge to help retail media networks maintain their growth trajectory while addressing these challenges.

Accept standardization when possible. While full integration across the network remains difficult, retail media players must invest in the baseline metrics and technical capabilities that sophisticated CPGs currently require. Ram Krishnan, CEO of North American beverages, said last year that in order to continue to acquire CPG Giant’s advertising business, the retail media network will need to “meta and other media they purchase,” including media they do on Google.

Pepsico evaluates your network based on several factors, including target capabilities, measurement clarity, creative flexibility, and, interestingly, API availability. “Not all retailers should be media companies,” his statement highlights the reality that small networks must make critical infrastructure investments to stay competitive.

Establish a core data infrastructure: Before introducing media attention-grabbing buzzibells and whistles, retailers need a solid data infrastructure. Lori Johnshoy, Head of Global Retail, Media Networks and CPG Industry Strategy at Liveramp, agrees to IAB’s highlights on the fragmented data ecosystem, which it says can limit the effectiveness of AI.

John Shoy, who previously helped launch the target round media platform, said this is why companies need trustworthy data collaboration partners. This is to help establish a cohesive data framework and bring all data points to a single accessible location. “In doing so, companies can maximize their AI investment potential, which is becoming increasingly essential in today’s competitive retail media network situation where demonstrating performance is important,” she says.

Exploring Real-Time Bid (RTB): RTB represents a promising technical solution to address the challenges of fragmentation while meeting the increasing demand for brand performance. As we looked into in a recent Forbes post, RTB enables automated impression-level bidding across multiple networks and provides a standardized flow of event-level data. While full RTB integration faces challenges from retailer walled gardens, even partial implementations can help address rising costs through more efficient allocations and provide the technical infrastructure needed for true cross-channel AI applications.

Improved AD relevance. Beyond automation and measurement, providing highly relevant advertising remains the most important success factor for retail media. Andreas Reiffen, CEO of AD Server Tech Company Pentaleap, said sponsored products account for around 80% of retail media revenue, but the CTR (click-through rate) for these ad types is only one-third of the results of organic products. This means that retailers are currently not offering relevant search ads to consumers, limiting the size of the retailer and brand’s advertising inventory. With more relevant paid and organic search results, Home Depot can surface 25 sponsored products per page, while many others show less than five. (Note, Pentaleap is my client.)

Future: Balanced Expectations

As the rapid growth of retail media begins to slow down, artificial intelligence appears not as a silver bullet, but as a strategic tool to address the industry’s most pressing challenges. From fragmented ecosystems and measurement difficulties to rising costs and integration complexity, AI offers targeted solutions that help retail media networks stay competitive.

However, successful implementation requires more than technical enthusiasm. A practical approach is needed to acknowledge the unique obstacles of the sector. Data standardization, privacy concerns and the need for a robust technological infrastructure. A thriving network is a network that views its strategic ability to carefully develop AI, not a simple fix for rushing deployment.

The future of retail media belongs to those who can gain a deeper understanding of technological innovation and the needs of advertisers. Use AI to create more relevant, measurable and efficient advertising experiences that bring true value to your brand and consumer.