73% of marketers say their AI marketing tools underperform expectations — not because the models are weak, but because the data feeding them is fractured beyond repair. That’s the uncomfortable truth hiding behind every “AI-powered” pitch deck in 2026. Identity resolution isn’t a nice-to-have layer anymore. It’s the thing determining whether your AI investment is a competitive edge or an expensive science project.
Abhishek Shukla, a data architect whose work has increasingly shaped enterprise martech conversations, has been pushing a framework that reframes identity resolution as infrastructure rather than a feature. Not a CDP add-on. Not a “we’ll get to it in Q3” line item. The foundation everything else sits on. If your brand is planning AI-driven personalization, predictive attribution, or agentic campaign management this year, his framework is worth understanding in detail — because most brand data architectures aren’t built for it.
Why AI Marketing Keeps Hitting a Wall
Here’s the pattern practitioners keep running into. You buy the AI platform. You feed it your customer data. The outputs are… fine. Not transformative. Not the 10x lift the vendor demo promised. Why?
Because most brands are still running identity systems built for a cookie-based, single-device, linear-funnel world that stopped existing years ago. A customer who clicks a TikTok ad on mobile, browses your site on a laptop, and completes a purchase in-store isn’t one identity in your system. They’re three or four fragmented profiles, each with partial signal, none of them talking to each other.
Feed an AI model fragmented identity data and it will optimize beautifully — for the wrong thing. It’ll double down on channels that look effective only because attribution can’t see the full path. It’ll personalize based on a sliver of behavior instead of the whole relationship. Garbage in, confident garbage out.
AI models don’t fix bad identity data — they amplify it, faster and at greater scale than any human marketer could.
This is exactly the gap Shukla’s framework targets. His argument, laid out in his original piece on identity resolution, is blunt: brands keep investing in AI capability while starving the identity layer that makes AI capability meaningful. It’s like buying a Formula 1 engine and bolting it onto a go-kart chassis.
What Shukla’s Framework Actually Proposes
Strip away the jargon and the framework rests on three architectural commitments most brands haven’t made yet:
- Deterministic-first, probabilistic-fallback matching. Hard identifiers (logged-in email, loyalty ID, CRM record) anchor the identity graph. Probabilistic signals (device fingerprints, behavioral patterns) fill gaps only when deterministic data runs out — not the other way around, which is how most legacy stacks are configured.
- Identity as a shared utility, not a departmental asset. Marketing, sales, and customer service teams often run separate, siloed views of “the customer.” Shukla’s model treats the identity graph as core infrastructure that every downstream system (ad platforms, CRM, personalization engines) queries from the same source of truth.
- Continuous resolution, not batch updates. Nightly batch syncs are too slow for real-time AI decisioning. The framework calls for near-real-time identity stitching so that an AI agent making a bid decision or a personalization call is working off current, not stale, data.
None of this is radical on paper. In practice, it requires brands to rebuild pipelines they’ve patched together for a decade. That’s the hard part — and why so few have actually done it.
The Cookie Deprecation Excuse Has Expired
For years, “we’re waiting on the cookieless transition to settle” was an acceptable reason to delay identity infrastructure work. That excuse doesn’t hold anymore. Google’s phased approach to third-party cookies in Chrome, combined with Apple’s ongoing ATT restrictions, means the deterministic-and-first-party-data world isn’t coming — it’s already here.
Brands that treated this as a future problem are now scrambling. Brands that built durable, consent-based identity graphs early are the ones whose AI tools are actually delivering. Our earlier coverage of anonymous audience marketing without cookies laid out exactly how this shift changes the mechanics of resolution — worth revisiting if your team is still leaning on third-party signal.
According to eMarketer research on first-party data strategy, the brands seeing the strongest returns on personalization spend are overwhelmingly the ones with consolidated, permission-based identity systems — not the ones with the flashiest AI vendor stack. The tool matters less than the foundation it sits on.
Attribution Breaks First When Identity Is Broken
If you want to see identity fragmentation’s damage in dollar terms, look at attribution. Marketing teams routinely misattribute conversions across channels because the systems tracking touchpoints can’t recognize that the same person interacted across five of them. This isn’t a rounding error — it’s often the difference between a channel looking like your top performer versus a money pit.
We’ve written before about how cross-system identity resolution fixes attribution gaps that CRM-native and ad-platform-native reporting simply can’t see on their own. Shukla’s framework leans into this directly: attribution accuracy is presented as a downstream symptom of identity health, not a separate problem to solve with a different tool.
The same logic extends to B2B. Buying committees don’t behave like single-threaded consumers — multiple stakeholders touch a deal, often anonymously, across different sessions and devices. Our analysis of AI attribution mapping for B2B buying groups shows how identity resolution has to account for group dynamics, not just individual journeys, to get ROI numbers brands can actually trust.
Where This Gets Operational: Agentic Marketing
Agentic marketing tools, the AI agents now handling bid adjustments, audience targeting, and even creative selection in near real time, are only as good as the identity signal they’re acting on. An agent deciding to increase spend on a “high-value segment” needs confidence that segment is coherent. If it’s actually three overlapping fragments of the same people counted multiple times, the agent is optimizing against a phantom audience.
This is the quiet risk in the current wave of agentic marketing architecture replacing static rule-sets. Autonomy without identity accuracy just means faster, more confident mistakes. Shukla’s framework essentially argues that agentic systems should be the last thing brands deploy, not the first — identity resolution has to be solved before autonomy is layered on top.
Vendor-side, this is already playing out. Amperity’s identity resolution work powering within-session personalization is a good example of the infrastructure-first approach in production: resolve identity accurately first, then let personalization and AI decisioning run on top of a stable graph.
Predictive Models Need Clean Identity Too
It’s not just attribution and agents. Predictive churn scoring, next-best-channel modeling, and creative performance forecasting all inherit whatever quality (or mess) exists in the identity layer beneath them.
Take churn prediction. A model trying to flag at-risk customers needs a complete behavioral history per person — support tickets, purchase cadence, engagement drop-off. If that history is split across three unlinked profiles, the model sees three partial, less alarming pictures instead of one clear warning sign. Our piece on predictive churn scoring gaps in CRM-native tools gets into exactly this failure mode.
Same story with next-best-channel engines replacing static media mix models: the “next best channel” recommendation is only trustworthy if the engine can see a person’s full cross-channel behavior, not a fragment of it.
Every AI marketing capability brands are racing to adopt — predictive, prescriptive, agentic — sits on the same load-bearing wall: identity. Crack that wall and everything built on top inherits the damage.
Building the Business Case: Where to Start
Practitioners reading this are probably thinking: fine, but where does budget actually go? A few concrete starting points, in rough priority order:
- Audit your current identity graph for duplication rate. Most brands are shocked to learn how many “unique” profiles are actually the same person split three or four ways.
- Consolidate consent and preference data alongside identity, not separately. Regulatory frameworks referenced by the FTC and the UK’s ICO increasingly expect consent to be traceable per identity record, not stored in a disconnected system.
- Move from batch to near-real-time resolution wherever AI decisioning depends on current data — bidding, personalization, agent-driven targeting.
- Treat the identity graph as shared infrastructure with a clear owner, not a byproduct of whichever team bought the CDP.
None of this is glamorous. It won’t show up in a highlight reel next to your flashiest AI campaign. But it’s the difference between AI tools that compound in value over time and ones that plateau fast because the data underneath never improves. For context on how measurement itself is evolving alongside identity, our coverage of deterministic vs. probabilistic attribution in modern MMM is a useful companion read for teams rebuilding measurement stacks in parallel.
Industry benchmarking from HubSpot on marketing data maturity echoes the same conclusion Shukla’s framework arrives at: the brands pulling ahead aren’t the ones with the most AI tools, they’re the ones with the cleanest, most unified data feeding a smaller number of well-chosen tools.
The Takeaway
Stop evaluating AI marketing vendors on model sophistication alone — audit your identity graph first, because that’s the ceiling on what any AI tool can actually deliver for your brand.
Frequently Asked Questions
What is identity resolution in marketing, in simple terms?
Identity resolution is the process of linking scattered data points, such as email addresses, device IDs, and behavioral signals, back to one real person or household, so brands have a single, accurate view of each customer instead of fragmented, duplicate profiles.
Why does identity resolution matter more now than in previous years?
Cookie deprecation and stricter mobile tracking rules have removed the easy third-party shortcuts brands relied on. At the same time, AI tools for personalization, attribution, and agentic decisioning require accurate, unified identity data to function correctly — without it, AI systems optimize against fragmented, misleading signals.
How is Abhishek Shukla’s framework different from a standard CDP approach?
Most CDPs treat identity resolution as one feature among many. Shukla’s framework positions identity as core infrastructure that every other system, from AI agents to attribution models, depends on, prioritizing deterministic matching first and continuous, real-time resolution over periodic batch updates.
What’s the first step for a brand with a fragmented identity system?
Start with an audit to measure duplication rates across your existing customer database. Understanding how badly identities are fragmented reveals the scale of the problem before you invest in new AI tools that would otherwise inherit the same fragmentation.
Does better identity resolution improve attribution accuracy?
Yes. Attribution models can only credit channels and touchpoints correctly if they recognize that different interactions belong to the same person. Poor identity resolution is one of the most common, and least discussed, causes of inaccurate attribution reporting.
FAQs
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 →