LinkedIn view-through conversions: count them in reporting or only click-based?

LinkedIn counts view-through conversions by default, so someone who saw a sponsored post, never clicked, and converted later still gets credited. Picture a B2B software company running single image ads where the platform reports a cost per lead half of what sales sees in the CRM. That gap is usually the view-through credit.

Some people argue it should stay on because long cycles need credit for influence. I'm not convinced that holds for budget decisions. If you judge bids and budget moves on that number, you end up paying for awareness that just overlapped with a branded search or a direct visit.

I'd keep view-through as a secondary column so it stays visible, but base bidding and budget calls on click-based conversions and reconcile against CRM opportunities every month. How do you handle it when the platform and the pipeline disagree?

Agreed on bidding, but the view window matters more than the flag. A 30-day window gives credit to people who were going to convert anyway, so I'd cut it to 1 day before arguing about the secondary column. Have you run a holdout to see what view credit actually adds?

It depends on your sales cycle, and honestly that matters more than the flag or the window. If deals close within a week, a 1-day view window is fine, since someone who saw an ad and converted later is rarely a real influence case. If your cycle runs 60 to 90 days, a 1-day window throws away almost all the credit you'd need to see anything, and what's left only counts people who were going to click anyway.

So I'd split it. Use the short window for bid and budget calls, because that's where you most want to avoid paying for overlap. Use the full-cycle window for the pipeline view, and reconcile it against CRM opportunities rather than trusting either number alone.

I'd be careful with the holdout. Geo holdouts on LinkedIn are expensive, and audiences are often too small to read cleanly within a quarter. Pausing one audience for a few weeks is cheaper, but it's noisy too.

I don't buy that a long cycle needs a long view window to "see anything." Window length says nothing about whether the ad caused the conversion. A 90-day window just gives you more room to overlap with branded search and direct visits, so the number gets bigger without getting any more causal. The lag is real, but the fix is to match view-exposed contacts to CRM opportunities and judge pipeline, not to trust a wider credit window. And on a 75-day cycle, pausing one audience for a few weeks won't tell you much, so I'd just be upfront that you can't read incrementality cleanly there.

Agreed that a wider window doesn't make the credit any more causal. But the window should come from your own data. Pull the lag distribution for clicked deals in the CRM. If most wins close within three weeks of first touch, a 90-day view window is mostly tail, and the extra credit goes to people who were already in motion. Say a SaaS team finds 80 percent of clicked-won deals close inside 21 days. A 30-day view window then covers nearly all the real lag, and anything past that is overlap.

Also check your match rate before you trust the reconciliation. If only part of the people who saw the ad can be matched to CRM contacts, your pipeline view is built on that slice. Compare close rates for the matched and unmatched groups. If they look very different, the gap between platform and CRM says more about who got matched than about what the ad did.

Set the view window at the 80th percentile of days from first touch to close for your clicked wins, not the platform default. Past that point you're mostly crediting people who were already on their way to buying, and the window stays tied to your own sales cycle instead of whatever the platform picked for you.