96% of marketers now use AI in their workflows. Only 44% trust the data feeding it. That 52-point gap isn’t an adoption problem. It’s an identity-resolution problem wearing an AI costume, and it’s quietly capping the ROI of every model, campaign, and dashboard built on top of it.
If you’ve watched a beautifully built AI tool underperform against its promise, this is probably why. Not the model. The identity layer underneath it.
The stat everyone quotes, and the one everyone skips
The 96% adoption figure gets cited constantly in trend decks and conference keynotes. It’s the headline that makes AI look inevitable. But the trust number rarely gets the same airtime, and that’s a mistake. When fewer than half of marketers trust the data quality behind their AI systems, you don’t have an innovation story. You have a foundation problem dressed up as a transformation story.
Think about what “trust” actually means operationally. It means a marketer looking at a churn prediction, a lookalike audience, or a next-best-action recommendation and thinking: I’m not sure this is built on accurate customer records. That hesitation shows up everywhere: in budget approvals that stall, in campaigns that get double-checked manually before launch, in leadership asking “where did this number come from” in every QBR.
An AI model can only be as confident as the identity graph beneath it. Garbage identity resolution doesn’t just produce bad outputs — it produces outputs that look plausible enough to act on, which is far more dangerous.
What identity resolution actually breaks down to
Identity resolution is the practice of stitching together every touchpoint a person generates, across devices, channels, and platforms, into one coherent profile. Sounds simple. It almost never is.
A customer browses on mobile Safari, clicks an Instagram ad on a different device, opens a marketing email on a work laptop, and eventually converts in-store using a loyalty card. In a mature identity system, that’s one person. In most stacks today, it’s three or four fragmented records, each with partial signal and no shared key.
AI doesn’t fix fragmentation. It amplifies it. A machine learning model trained on fractured identity data will confidently recommend budget shifts, audience expansions, or creative variants based on a customer who, statistically speaking, doesn’t fully exist. The model isn’t wrong. The inputs are.
Why this gap widened as AI adoption accelerated
Here’s the uncomfortable part: AI adoption outpaced data governance almost everywhere. Marketing teams bought or built AI tools faster than they cleaned up the identity infrastructure those tools depend on. It’s the classic build-the-roof-before-the-foundation problem, just automated and scaled.
Three forces accelerated this:
- Platform fragmentation. The average enterprise marketing stack now spans a dozen or more tools, per HubSpot’s own research into martech sprawl, each with its own identifier logic and none of them talking to each other cleanly.
- Cookie deprecation pressure. As third-party identifiers faded, teams rushed to stitch together first-party signals without always validating match quality first.
- Vendor incentive misalignment. Most AI marketing tools are sold on capability, not on data readiness. Nobody in a sales demo says “by the way, this only works if your identity graph is clean.”
Our previous coverage on why AI agents need clean data first makes a similar point from the master data management angle: agentic systems don’t just need more data, they need trustworthy data, and those are very different requirements.
Root cause one: duplicate and orphaned records
Ask any ops lead how many duplicate customer records live in their CRM and watch them wince. Duplicate records aren’t a cosmetic issue. They split behavioral history across multiple profiles, which means your AI model sees half a customer’s journey and treats it as a whole one.
Orphaned records are worse in a different way. These are touchpoints, an ad click, a form fill, a support ticket, that never get matched to any known identity at all. They sit in the data warehouse as noise. Feed enough noise into a propensity model and you get confident-sounding predictions with no real signal behind them.
This is exactly the diagnostic ground covered in identity resolution and governance: buying another resolution tool doesn’t help if nobody owns the standards for what counts as a valid match.
Root cause two: siloed consent and permission data
Privacy regulation added a second layer of complexity that most identity conversations skip. It’s not enough to know that Profile A and Profile B are the same person. You also need to know what that person consented to, in which jurisdiction, and under which policy version.
Marketers operating under FTC guidance and, for UK/EU operations, ICO requirements, are increasingly required to prove not just that identity resolution happened, but that it happened within consent boundaries. An AI system that merges identities without respecting consent scope isn’t just inaccurate. It’s a compliance liability.
This is where a lot of “we trust our AI” claims fall apart under audit. The model works. The governance behind it doesn’t.
Root cause three: attribution built on stitched-together guesses
Here’s where the ROI conversation gets real. If your identity graph is shaky, your attribution model is shaky, and if your attribution model is shaky, every budget decision downstream inherits that uncertainty.
Consider the CRM-to-media-platform handoff. A prospect’s last-touch data point might live in a paid social platform, while their actual purchase happened three weeks later through an entirely different channel, tracked in the CRM under a slightly different email variant (say, a personal address versus a work one). Without deterministic matching between those systems, attribution defaults to modeled guesses, and AI-driven budget allocation tools will happily optimize toward those guesses as if they were fact.
Our piece on how identity resolution meets CRM attribution digs into exactly this failure mode, and it’s one of the more fixable root causes on this list, provided teams invest in deterministic matching rather than relying purely on probabilistic stitching.
Bad identity resolution doesn’t just cost accuracy. It costs budget, because AI systems reallocate spend toward whatever pattern the flawed data suggests is working.
So what actually closes the gap?
There’s no single tool that fixes this. Anyone pitching a one-click identity resolution solution is selling you the same false confidence that created the problem. But there is a sequence that works.
Audit before you automate. Before deploying or expanding any AI marketing tool, run a match-rate audit. What percentage of your customer records are confidently resolved versus probabilistically guessed versus completely orphaned? Most teams have never actually measured this. It’s an uncomfortable number, and it’s the single most useful diagnostic you can run this quarter.
Separate deterministic from probabilistic matches in your reporting. Not all “resolved” identities are equal. A match based on a shared logged-in email is fundamentally more reliable than one based on device fingerprinting and behavioral similarity. AI dashboards rarely make this distinction visible, which is part of why trust erodes. Marketers see one confidence score, not the matching methodology behind it.
Assign governance ownership, not just tool ownership. Someone needs to own the standards: what counts as a valid identity link, how consent flows through merges, how conflicts get resolved when two systems disagree about who a customer is. Without a named owner, this work falls through the cracks between IT, marketing ops, and legal.
Pressure-test vendor claims about “AI-ready data.” When evaluating new platforms, whether it’s a CDP, a CRM add-on, or a full conversation-first CRM approach, ask specifically how the platform handles identity conflicts, not just how it handles identity matches. The interesting failure cases are always in the conflicts.
Treat segmentation tools with the same scrutiny. AI-powered auto-segmentation is only as trustworthy as the identity data feeding it. The vetting process outlined in coverage of AI auto-segmentation for cause-marketing data applies broadly: segment quality is a downstream symptom of identity quality, not a separate problem.
The uncomfortable ROI math
Here’s the framing that tends to get budget conversations moving. According to eMarketer’s ongoing tracking of martech investment, spend on AI-driven marketing tools continues climbing year over year. Meanwhile, the trust gap this article opened with hasn’t meaningfully closed. That means a growing share of marketing budget is being deployed on top of infrastructure that half the industry doesn’t fully trust.
That’s not a technology risk. That’s a balance-sheet risk. Every dollar spent scaling an AI tool before fixing identity resolution is a dollar at greater risk of misallocation. Fix the foundation first, and the same AI tools suddenly perform closer to their advertised ceiling, because they’re finally working with inputs worth trusting.
The gap between 96% adoption and 44% trust won’t close through better models. It closes through better plumbing: cleaner identity graphs, visible match confidence, and governance that someone actually owns. Start with the audit, not the next tool purchase.
FAQs
Why do marketers trust AI adoption more than AI data quality?
Adoption is easy to measure and easy to mandate: a team either uses a tool or it doesn’t. Data quality is harder to audit, often invisible until a campaign underperforms, and rarely owned by a single team, which is why trust lags far behind usage.
What is identity resolution in marketing, in plain terms?
It’s the process of matching all the different digital footprints one customer leaves, across devices, emails, and platforms, into a single accurate profile. Poor identity resolution means your systems think one customer is actually two or three separate people.
How can a brand measure its identity resolution quality?
Run a match-rate audit that separates deterministic matches (like shared logged-in emails) from probabilistic ones (like device or behavioral similarity), then check what percentage of records remain orphaned or unmatched entirely.
Does fixing identity resolution require buying new software?
Not necessarily. Many gaps come from governance failures, no one owning match standards or consent rules, rather than a missing tool. Governance and process fixes often deliver more improvement than another platform purchase.
What’s the business risk of ignoring the identity-resolution gap?
AI tools built on fragmented identity data will confidently misallocate budget, misfire attribution, and generate outputs that look accurate but aren’t. The risk compounds as AI spend increases without a corresponding investment in data foundations.
Top Influencer Marketing Agencies
The leading agencies shaping influencer marketing in 2026
Agencies ranked by campaign performance, client diversity, platform expertise, proven ROI, industry recognition, and client satisfaction. Assessed through verified case studies, reviews, and industry consultations.
Moburst
-
2
The Shelf
Boutique Beauty & Lifestyle Influencer AgencyA data-driven boutique agency specializing exclusively in beauty, wellness, and lifestyle influencer campaigns on Instagram and TikTok. Best for brands already focused on the beauty/personal care space that need curated, aesthetic-driven content.Clients: Pepsi, The Honest Company, Hims, Elf Cosmetics, Pure LeafVisit The Shelf → -
3
Audiencly
Niche Gaming & Esports Influencer AgencyA specialized agency focused exclusively on gaming and esports creators on YouTube, Twitch, and TikTok. Ideal if your campaign is 100% gaming-focused — from game launches to hardware and esports events.Clients: Epic Games, NordVPN, Ubisoft, Wargaming, Tencent GamesVisit Audiencly → -
4
Viral Nation
Global Influencer Marketing & Talent AgencyA dual talent management and marketing agency with proprietary brand safety tools and a global creator network spanning nano-influencers to celebrities across all major platforms.Clients: Meta, Activision Blizzard, Energizer, Aston Martin, WalmartVisit Viral Nation → -
5
The Influencer Marketing Factory
TikTok, Instagram & YouTube CampaignsA full-service agency with strong TikTok expertise, offering end-to-end campaign management from influencer discovery through performance reporting with a focus on platform-native content.Clients: Google, Snapchat, Universal Music, Bumble, YelpVisit TIMF → -
6
NeoReach
Enterprise Analytics & Influencer CampaignsAn enterprise-focused agency combining managed campaigns with a powerful self-service data platform for influencer search, audience analytics, and attribution modeling.Clients: Amazon, Airbnb, Netflix, Honda, The New York TimesVisit NeoReach → -
7
Ubiquitous
Creator-First Marketing PlatformA tech-driven platform combining self-service tools with managed campaign options, emphasizing speed and scalability for brands managing multiple influencer relationships.Clients: Lyft, Disney, Target, American Eagle, NetflixVisit Ubiquitous → -
8
Obviously
Scalable Enterprise Influencer CampaignsA tech-enabled agency built for high-volume campaigns, coordinating hundreds of creators simultaneously with end-to-end logistics, content rights management, and product seeding.Clients: Google, Ulta Beauty, Converse, AmazonVisit Obviously →