Most ecommerce founders we onboard arrive with the same dashboard: impressions trending up, follower count trending up, the occasional viral post. Then they ask the question that breaks every reporting tool we've seen: "Where's the pipeline?"
Founder LinkedIn attribution is broken at almost every brand we audit — not because the data isn't there, but because nobody is collecting the four signals that actually matter. Likes and impressions are the easiest to count, which is exactly why they dominate the dashboard. They're also the worst predictors of inbound revenue.
After running content for 47 ecommerce founders over the last 24 months, here's the attribution model we now use to separate noise from pipeline.
Why the Default LinkedIn Dashboard Lies
LinkedIn Analytics gives you four numbers by default: impressions, reactions, comments, and shares. Three of those four are weakly correlated with pipeline. Impressions in particular have almost zero correlation — we've seen 80,000-impression posts generate zero qualified inbound and 4,200-impression posts generate three discovery calls.
The reason: LinkedIn's feed pushes content to dormant connections and lookalike audiences first to test relevance. Most of those people will never be a customer. A post hitting 80K impressions usually means the algorithm found a broad emotional hook (career advice, layoffs, motivation), not a buying audience.
Pipeline lives in a much smaller, much harder-to-see signal layer.
The 4-Layer Founder Attribution Model
We track four layers per post. Anything below layer 2 is noise.
Layer 1: Reach quality, not reach volume
We don't care about impressions. We care about ICP impressions — the percentage of impressions reaching your defined ideal customer profile.
LinkedIn doesn't give you this directly. We approximate it by exporting the post's "top viewers" sample (visible inside Creator analytics for posts with 1,000+ impressions) and scoring against the founder's ICP filter (industry, seniority, company size). If less than 35% of viewers match ICP, the post is a vanity hit regardless of impression count.
Benchmark across our roster: healthy founder posts run 42-58% ICP-matched reach. Viral posts almost always drop below 20%.
Layer 2: Profile-view conversion
This is the most underused number on LinkedIn. The path to inbound is almost always: post seen → profile viewed → DM, connect, or website click.
We measure profile views per post, not weekly. Healthy founder posts drive 0.4-1.2% profile-view-per-impression. Below 0.2% means the hook landed but the writing didn't earn curiosity about the author.
When this number drops, the fix is rarely the post — it's usually that the founder pillared themselves into a topic that doesn't pay them back (industry commentary instead of category authority).
Layer 3: DM and connection-request composition
Volume of DMs is meaningless. Composition is everything. We tag every inbound DM into one of five buckets:
- Buyer — fits ICP, has a stated problem
- Peer — same industry, networking intent
- Vendor — selling to the founder
- Recruiter — hiring intent
- Random — unclassified
For founders running a B2B angle (agencies, SaaS, services), we want 40%+ buyer-tagged DMs. For pure ecom DTC founders using LinkedIn for partnership and PR pipeline, we want 35%+ peer + 15%+ buyer.
If buyer DMs sit below 15% for three weeks, the content pillar mix is wrong — almost always too much "thought leadership" and not enough specific problem articulation.
Layer 4: Self-reported attribution on calls
The single most reliable attribution signal is also the lowest-tech: ask every booked call where they found you.
We require every founder we work with to add one field to their discovery call template: "How did you hear about me?" Free text. We tag responses weekly into LinkedIn-post (with post ID if they mention one), LinkedIn-profile (general), referral, podcast, search, or other.
Of all attribution signals, this is the one that closes the loop. We've seen brands try UTM-tagged links, dedicated landing pages, custom Calendly URLs per post — all degrade content performance because they break native engagement. The discovery-call question costs zero reach and produces the cleanest data.
What "Good" Looks Like at 90 Days
A founder running this attribution model from day one — across 3-4 posts per week, on a defined pillar — should hit by day 90:
- 42%+ ICP-matched reach on average across the period
- 0.5%+ profile-view-per-impression average
- 5-12 inbound DMs/week, with 30%+ buyer-tagged
- 2-5 discovery calls/month self-attributed to LinkedIn
If those numbers are under, the issue is almost never the founder's writing ability. It's pillar drift, ICP mismatch, or posting cadence (more on that in our 18-hour gap rule analysis).
The Reporting Cadence That Actually Works
We send our founder clients one report per month. It's a single page. It contains:
- ICP-matched reach % (vs. last month)
- Profile views (vs. last month)
- DM composition % by bucket
- Self-attributed calls and their stage in pipeline
- The 3 posts that produced the most layer 2-4 signal — and why we think they did
That's it. No impressions chart. No follower-growth graph. No engagement-rate trend.
The report fits on one page because the goal of founder LinkedIn isn't to grow on LinkedIn — it's to grow off LinkedIn. Every metric we track is a leading indicator for off-platform action.
Common Attribution Mistakes We See
A few patterns from auditing other agencies' setups:
- Counting comments as pipeline — most comments come from peers and content reciprocity networks, not buyers
- Using "engagement rate" as the headline KPI — engagement rate rises when reach falls, so this number can climb while pipeline collapses
- Reporting on follower growth — followers gained in months 1-6 of a content push are mostly low-intent. Follower-to-DM ratio is more useful.
- Attributing every booked call to "LinkedIn" because the founder posts on LinkedIn — without the discovery-call question, this is a guess
FAQ
How long before founder LinkedIn shows pipeline signal? With our roster, the median is week 9. Anything before week 6 is usually warm-network conversion, not algorithmic reach.
Should I build a personal brand or a company brand on LinkedIn? For ecom founders under $20M revenue, founder brand outperforms company page reach by roughly 8-15x in our tracking. Company pages become useful for retargeting and credibility, not pipeline.
What posting frequency works best for attribution? 3-4 long-form posts per week, with at least 18 hours between posts. Daily posting suppresses individual post reach and dilutes attribution.
If you want help building this attribution model into your own founder content workflow, let's talk. We run this system for our entire client roster — happy to share the dashboard template either way.