Table of Contents
- What Is Identity Fragmentation?
- Why One Buyer Becomes Many Records
- How One Buyer Looks to a Platform
- Cross-Device Identity: Why Recognizing the Same Buyer Is Hard
- What Fragmentation Actually Costs You
- Why This Is Worse in 2026, Not Better
- How Platforms Try to Solve This, and Where They Fall Short
- What to Ask Your Platform About Identity
- Key Takeaways
- Frequently Asked Questions
You’re no doubt familiar with this scenario: A potential customer researches a purchase on their phone at lunch, compares options on a laptop that evening, and opens your company’s app over the weekend. To you, that’s one buyer moving toward a decision across three sessions, but to an ad platform, it can look like three unrelated people.
What appears to be a tracking-accuracy issue is actually a recognition problem. When an ad platform can’t connect a person’s activity across the purchase journey, the effects of fragmented identity show up downstream, affecting everything from audience size and frequency to suppression, bidding, and measurement.
Here’s what identity fragmentation is, why it’s getting worse, and what it costs you.
What Is Identity Fragmentation?
Identity fragmentation occurs when a single person is represented as multiple unconnected records across devices, browsers, environments, and sessions, so a platform can’t tell that it’s seeing the same buyer twice. Although often lumped in with the decline of third-party cookies, the problem actually predates these restrictions, since cookies are just one type of identifier, and losing them is only one source of fragmentation.
Why One Buyer Becomes Many Records
Several factors can cause the same customer to appear as multiple platform records:
Multiple Devices, Multiple Sessions
Moving from phone to laptop to tablet is routine, and each device starts out with a separate identity unless something connects them.
Different Environments Don’t Share Identifiers
This is the cause advertisers tend to underestimate. Web, in-app, and connected TV (CTV) rely on different identifier systems that don’t automatically connect. A browser may use cookies and publisher IDs; an app may use a mobile advertising ID; and CTV can rely on device, household, or authenticated signals. Someone who sees a CTV ad, researches the product on their phone, and later opens an app has crossed three systems with no bridge between them. Each records the activity separately.
Identifiers Expire, Reset, and Get Blocked
Identifiers don’t last indefinitely. Cookies expire or get cleared, mobile advertising IDs reset, and browser settings can restrict tracking, so someone recognized in one session can appear under a new identifier later.
Most Browsing Is Logged Out
Most sessions have no login and no authenticated signal to anchor activity to a known account, so ad platforms lean on weaker device and browser signals to guess who they’re looking at. Without that anchor, the same person is harder to recognize across later sessions.
How One Buyer Looks to a Platform
The logged-in visit is the most revealing point in the journey: the ad platform finally has a strong identity signal, but the buyer’s earlier anonymous activity still remains disconnected.
| Touchpoint | What Identifier Is Available | What the Ad Platform Sees | What It Can’t Connect |
| The buyer researches on mobile web. | A browser or publisher identifier is available. | Anonymous category research is recorded. | It can’t tell whether this buyer appeared elsewhere. |
| The buyer compares options on desktop. | A different browser identifier is available. | Another anonymous comparison session is recorded. | It can’t connect the session to the mobile research. |
| The buyer opens the retailer’s app. | A mobile advertising ID is available. | Activity is tied to that device. | It can’t connect the app activity to earlier web sessions. |
| The buyer visits while logged in. | An authenticated account identifier is available. | A known account is recognized. | It can’t automatically connect the earlier anonymous activity. |
| The buyer completes the purchase. | A transaction or account signal is available. | A completed conversion is recorded. | It can’t connect every earlier touchpoint that influenced the purchase. |
Cross-Device Identity: Why Recognizing the Same Buyer Is Hard
Cross-device identity is the ability to recognize that separate identifiers and sessions across devices, browsers, and digital environments belong to the same person. Confidence in this varies, however: a shared login or hashed email is a reliable match, while a match built on looser overlap is closer to a good guess. That range, from solid to speculative, is what makes cross-device identity genuinely difficult.
What Fragmentation Actually Costs You
Disconnected records lead to flawed campaign decisions in four common ways:
You Pay to Reach the Same Person Repeatedly
When identity is fragmented, what appears on your dashboard as broad, incremental reach is often just aggressive ad repetition targeted at the same person. This inflates your effective customer acquisition cost while annoying the very consumer you’re trying to convert.
Your Audience Sizes Are Wrong
If the same buyers appear under multiple identifiers (picture one household with six or seven distinct digital identifiers across different devices and platforms), then once again, your unique audience looks much larger than it really is. This distorts any reach and frequency estimates and, worse, the planning built on them.
Suppression Fails
Suppression is designed to stop customer acquisition campaigns from wasting spend on existing buyers, but it only works when the ad platform recognizes those who already converted.
When an existing customer keeps appearing as a fresh prospect on another device or browser, the platform will treat them as a brand-new prospect, continuing to serve top-of-funnel conversion ads to someone who just completed a purchase.
Measurement Understates What Worked
Disconnected tracking forces attribution models to assign full conversion credit to the last recognizable touchpoint. A user may discover your product via an upper-funnel native ad on their mobile phone, research it on a tablet, and finally convert days later on a desktop search link, but because the early touchpoints live on separate identity fragments, the converting session gets disproportionate credit.
Why This Is Worse in 2026, Not Better
The current cookie situation is certainly confusing — Google confirmed in 2025 that it would keep Chrome’s existing cookie controls, rather than introduce the new standalone prompt it had proposed the year before; Safari blocks third-party cookies by default; and Firefox restricts cross-site tracking cookies, so across browsers, these three different approaches make cookie-based recognition less consistent — but cookies are only one part of the problem.
The bigger issue is that buyers are constantly moving across web, apps, CTV, retail media, search, social, and publisher environments that rely on different identifiers, giving each ad platform only a partial view of the journey. Consent gaps and ongoing identifier churn create additional breaks in recognition, which exacerbates the problem.
How Platforms Try to Solve This, and Where They Fall Short
Even sophisticated identity resolution is only as complete as the footprint it’s built on. Most ad platforms try to reconcile fragmented records into a single view by matching identifiers when a user logs in, when a hashed email is available, or when overlapping signals are strong enough to link sessions with confidence. Since each can only resolve identity across properties it has a direct relationship with, though, activity outside that footprint can’t be seen or connected.
What to Ask Your Platform About Identity
When evaluating identity capabilities, focus on how the platform resolves users and where its visibility ends:
- How do you resolve identity across browsers, devices, and environments?
- What inventory and publisher footprint can you make those connections across?
- What happens when an identifier expires, resets, or becomes unavailable?
- Is resolved identity applied to measurement, suppression, and targeting?
- What share of your addressable reach is authenticated versus inferred?
Key Takeaways
Identity fragmentation is a recognition problem, not a tracking problem, because the ad platform fails to realize it has already seen the same buyer it’s currently targeting as a new possibility. It inflates perceived reach while hiding what really drove the conversion, while disconnected environments, consent gaps, and identifier churn keep making identity resolution harder, regardless of third-party cookie policy.
Frequently Asked Questions
Is identity fragmentation the same as signal loss?
No. Signal loss means an identifier or data point is unavailable. Identity fragmentation means available records can’t be reliably connected to the same person. Signal loss can contribute to fragmentation, but the two aren’t the same.
Did Google keeping third-party cookies fix fragmentation?
No, because fragmentation extends well beyond Chrome. Safari blocks third-party cookies, Firefox restricts cross-site tracking cookies, apps and CTV use different identity systems, and many users remain logged out while browsing.
How do I know if fragmentation is affecting my campaigns?
Look for inflated reach, customers still receiving acquisition ads after converting, inconsistent frequency, and measurement that gives too much credit to the last recognizable touchpoint.
Is fragmentation only a problem for large advertisers?
No. Any advertiser reaching buyers across multiple devices, browsers, publishers, or platforms can encounter it. Scale affects how much budget is exposed, not whether fragmentation distorts audience sizing, suppression, frequency, and measurement.
