Comparing CRM vs ad platform conversions is the fastest way to find out whether your reporting means anything, and the answer in my accounts is duller than this topic usually gets written up. The platform typically reports about 10% more than the CRM records. A normal week looks like 50 leads on the platform and 45 in the system.
That is well short of the multiples this genre tends to quote as standard. It is also the number I would want you to check rather than assume.
This is for you if marketing and sales are reporting different lead counts and nobody has sized the difference.
What CRM vs ad platform conversions are each counting
They count different events, at different moments, under different rules.
Your ad platform counts a tag firing. Something happened on the page that it was told to record, and it credits that to an interaction it served, inside a window it chose.
Your CRM counts a record arriving. A lead reached the system, usually with a name attached, often after passing through an integration and sometimes after a human looked at it.
Between those two moments sit spam filters, failed integrations, deduplication, and anybody who deleted an obvious junk submission. Every one of those removes records from the second count without touching the first.
So a gap is the normal condition. Its size is the useful information.
Why the usual claim about the gap is overstated
Most writing on this subject implies the platform inflates by a large multiple as a rule. My accounts do not show that.
Around 10% is the figure I see repeatedly across lead generation clients with tracking set up properly. Some run tighter. A small number run at two or three times, and when they do, there is a specific cause rather than a general truth about platforms.
I think the exaggeration has a simple explanation. A 10% discrepancy does not justify buying anything. A 300% one does, and most of the published material comes from companies selling the remedy.
That is not an accusation of dishonesty. It is a reason to measure your own rather than adopt somebody else’s number, especially when the number is being used to sell you something.
The exception worth knowing is that ecommerce behaves differently from lead generation here. View-through conversions and long windows do more work in a purchase funnel, so purchase-side gaps are genuinely wider.
There is one published comparison worth holding beside your own figure. Seer Interactive tested Meta’s incremental attribution setting across $1.05 million of ad spend on six accounts in April 2025. Meta reported 87% of conversions as incremental where GA4 put the same accounts at 67%, a twenty percentage point gap.
That is two measurement systems rather than a platform against a CRM, so it is not the same comparison. It is useful anyway, because it shows how far two honest systems can drift while both work correctly. Six accounts is a small sample and the study says so.
The four causes of the CRM vs ad platform conversions gap
In the order they usually contribute.
Spam and duplicates. The platform counts every submission. Somebody in the business deletes the obvious junk, and those records never reach the CRM. This is the biggest single component in most accounts, and it means your real cost per usable lead is higher than reported.
Integration failures. A form submits, the tag fires, and the handoff to the CRM silently drops it. Nothing errors. The lead simply is not there.
Timing. Attribution windows credit conversions to the week of the click rather than the week of the enquiry, so a like-for-like monthly comparison is not actually like for like.
View-through. Somebody who never clicked gets counted by the platform and has no corresponding record anywhere, because from your CRM’s point of view they were never a lead.
How to size yours properly
Match the periods before you compare anything, because most reported gaps shrink once you do.
Pull the same date range on both sides. Leave out the most recent week, so conversion lag does not make the platform look wrong when it is only slow. Then count the same event on both sides, which usually means the specific form or forms the campaign points at rather than all enquiries from every source.
Write down the result as a percentage. That single number is worth more to you than any published benchmark, because it describes your account rather than someone’s customer sample.
Then repeat it monthly for three months. Three data points tell you whether it is stable, and stability is the thing you actually care about.
Why stability beats accuracy here
A gap you can predict is not a problem. A gap that moves is.
If the platform reports 10% above the CRM every month, you can work with that indefinitely. Steer campaigns on platform data, report the CRM figure upward, and apply the correction when you need to translate between them.
When that stable 10% becomes 40% in a month, something broke. Usually an integration, usually after a site change, and usually nobody was told. That movement is a genuinely good alarm, and it only works if somebody wrote down the baseline.
Which is the argument for measuring it even when everything looks fine. The baseline has no value on the day you record it and considerable value six months later.
What this changes about your cost per lead
Every efficiency figure you quote sits on one of these two counts, and most people do not know which.
If your cost per lead is calculated on platform conversions, it includes spam and duplicates, so it flatters you. Recalculate on CRM records and it will rise. That higher number is the honest one, and it is the one your sales team already believes.
I would rather a client used the CRM figure and knew it was conservative than used the platform figure and thought it was exact. The second is how budgets get set on numbers nobody can defend later.
What to change this week
Three steps.
Pull one month from both systems, matched on period, and write the gap down as a percentage. That takes about half an hour and you will refer to it for years.
Then check your form-to-CRM integration actually fires, by submitting a test enquiry yourself and confirming it arrives. Integration failures are invisible from both dashboards.
Then decide which of the two numbers the business reports, and say so out loud, so nobody is quietly quoting the other one in another meeting.
What conversion numbers actually measure is the mechanism underneath all of this, platform reported conversions explains why the platform side counts what it counts, the four ways conversion tracking goes wrong covers the failure modes behind an unstable gap, and lead quality is what the deleted records were really telling you. Fixing a broken handoff is usually Google Ads work and website work at the same time. If you want yours sized and explained properly, book a teardown.