If you’ve been running LinkedIn ads for more than a minute, you’ve probably seen “Conversion API” pop up in a LinkedIn blog post or a vendor pitch deck. And if you’re like most of the marketers we talk to, your reaction was somewhere between “I should look into that” and “one more thing to figure out.”
Here’s why it’s worth actually figuring out: CAPI is the single biggest lever most B2B advertisers aren’t pulling when it comes to data quality. Not a nice-to-have. Not a 2026 priority. Something that’s almost certainly affecting your reporting and your optimization right now.
Let me walk through what it actually is, why it matters specifically for B2B, and what you need to know before you start implementing.
The Short Version
CAPI, or LinkedIn’s Conversions API, is a server-to-server connection between your systems and LinkedIn’s ad platform. Instead of relying on a browser-based pixel to report conversions, your server sends that data directly to LinkedIn.
That one-sentence change has a lot of downstream implications.
Impactable is a certified LinkedIn CAPI implementation partner.
We’re one of the few agencies recognized by LinkedIn for implementing the Conversion API at scale — with the certified team and infrastructure to handle end-to-end CAPI setup, data mapping, and ongoing measurement management. If you’re dealing with signal loss or attribution gaps on LinkedIn, we can help you fix the foundation.
Talk to us about CAPI →Why the Insight Tag Alone Isn’t Enough
LinkedIn’s standard tracking tool is the Insight Tag — a piece of JavaScript you install on your website. When a visitor converts, the tag fires in their browser and reports back to LinkedIn.
This works fine when the browser cooperates. The problem is that browsers increasingly don’t cooperate.
Ad blockers intercept the tag before it loads. Apple’s Intelligent Tracking Prevention (ITP) caps first-party cookies at 7 days on Safari. Privacy browsers like Brave and Firefox block third-party scripts by default. Any of these, silently, cuts the attribution chain before LinkedIn ever sees a conversion.
For most consumer audiences, this is an annoying data quality issue. For B2B audiences, it’s much worse. The people you’re paying $10–$15+ per click to reach — senior buyers, IT decision-makers, VPs and above — are disproportionately likely to use ad blockers. Estimates put ad blocker usage at 30–50% in these segments. That means for a meaningful chunk of your most qualified traffic, the Insight Tag sees nothing.
There’s also the sales cycle problem. LinkedIn extended their attribution window to 365 days specifically because B2B deals take time. But if a buyer clicks your ad in month one and closes in month seven, and ITP expired their cookie on day seven, LinkedIn never connects those dots. The campaign that drove the deal never gets credit.
CAPI bypasses all of this. Your server fires the event. No browser involved.
What You Can Actually Send Through CAPI
This is where it gets useful for B2B specifically. You’re not limited to website events. Through CAPI, you can send:
- Online conversions — form fills, demo requests, downloads, page-specific events
- Offline conversions — phone calls, in-person event attendance, deals closed offline
- CRM-connected events — MQL updates, SQL stage changes, opportunity creation, closed-won
- Custom events — anything specific to your product or business that indicates intent
The CRM piece is a big deal. Most accounts we see are only tracking form fills — which is a proxy for pipeline, not pipeline itself. When you close the loop and send “deal closed” events from your CRM back to LinkedIn, the algorithm learns what your actual revenue-driving audience looks like. That changes who it targets, not just what it reports.
How the Matching Works
When someone clicks a LinkedIn ad, LinkedIn appends a Click ID — called li_fat_id — to the landing page URL. Your server captures that ID when the visitor converts and sends it back with the CAPI event.
This is called deterministic matching, and it’s the key to accurate attribution. LinkedIn can match the conversion to the specific click that caused it, with accuracy above 95%.
The alternative — when you don’t pass the li_fat_id — is probabilistic matching using email addresses. This fails regularly in B2B, where someone might submit your form with their work email but have a personal Gmail on their LinkedIn profile. Same person, different emails, no match.
Make sure your CAPI setup captures and passes the li_fat_id. Without it, you’re getting probabilistic matching at best, and your match rates could sit in the 40–60% range instead of 90%+.
CAPI vs. Insight Tag Is Not Either/Or
You don’t replace the Insight Tag with CAPI. You run both.
The Insight Tag is still valuable — for building retargeting audiences, website demographic reporting, and basic behavioral data. CAPI handles conversion accuracy. When a conversion comes in through both channels simultaneously, LinkedIn de-duplicates, so you’re not double-counting.
What you’re doing by running both is filling in the gaps. The Insight Tag catches what it can. CAPI catches what the Insight Tag missed. The result is a more complete picture of what your campaigns are actually doing.
LinkedIn’s own data shows advertisers using CAPI see roughly a 31% increase in attributed conversions compared to Insight Tag-only tracking. To be clear about what that number means: it’s not 31% more conversions happening. It’s 31% more conversions that LinkedIn can now see and claim credit for — ones that were always occurring but couldn’t be tracked. As Justin put it: almost a third of your results were going uncounted.
For more on how to evaluate your conversion data holistically, our guide on measuring LinkedIn ads performance beyond conversions goes deep on what signals to track and how to triangulate them.
Implementation Paths
There are a few ways to set this up, depending on your tech stack and resources:
Native CRM integrations — LinkedIn has pre-built connectors with HubSpot, Salesforce, and a handful of others. If you’re on one of these platforms, this is usually the fastest path to getting data flowing.
Google Tag Manager (server-side) — Popular approach. One important gotcha: if your client-side GTM is being blocked by ad blockers, your server-side container receives nothing. The whole point of CAPI is bypassing browser tracking issues, so make sure the li_fat_id is being captured at the server level, not relying on client-side data.
Direct server integration — Your engineering team sends events from the backend. Most control, highest accuracy, requires dev resources.
Third-party tools — Platforms like Zapier, Supermetrics, and others offer managed integrations that handle API version changes on your behalf.
The right path depends on your setup. The wrong path is doing nothing, which is where most B2B advertisers still are.
The Executive Framing
At some point someone senior is going to ask about LinkedIn ROI. If your data is based entirely on Insight Tag tracking and 30–50% of your audience was invisible to it, your CPL numbers aren’t directionally off — they’re built on a flawed foundation.
That works against you in both directions. It might mean you’re underreporting LinkedIn’s contribution, making it look like a weaker channel than it is. Or it might mean you’ve been over-optimizing toward audiences that convert in the browser while missing the higher-intent buyers who don’t.
Getting CAPI in place isn’t just a tracking project. It’s the infrastructure that makes it possible to have a credible conversation about what LinkedIn is actually doing for the business — which is a conversation your CFO will eventually ask you to have.
We cover more of this attribution framing, including how to reconcile LinkedIn data with GA4 and CRM numbers, in our post on how to measure LinkedIn ads performance.
Before You Do Anything: Audit First
Before jumping into implementation, check where you stand:
- Is your Insight Tag firing on all pages? Check the domain status in Campaign Manager under Analyze > Insight Tag.
- Are all your conversion rules attached to active campaigns? Unattached rules collect nothing.
- Is li_fat_id being captured when visitors land from LinkedIn? This is the single most important setup detail for CAPI accuracy.
- What’s your attribution window set to? If your sales cycle runs 6 months but your window is 30 days, you’re missing a lot.
If you want a structured way to work through this, our LinkedIn Ads Survival Kit includes a full tracking audit template. We’re also a CAPI-certified LinkedIn Marketing Partner — so if you’d rather have someone do it with you, book a discovery call.
