Google Ads showing conversions but no leads
If Google reports more conversions than you have enquiries, something is being counted that is not a customer. The useful question is which action produces those conversions, and whether your campaigns are being optimised toward it.
Say the report from your agency shows 47 conversions this month.
The natural assumption is that those are enquiries, bookings or sales. But without knowing what is being counted, how would you know what those 47 things actually are?
They could be 47 site visits, if a page view is being recorded as a conversion.
The number itself tells you nothing until you know which action produces it. That is the question an independent check answers.
Six things that get counted instead of a customer
Notice that none of these requires anybody to have acted in bad faith. Every one is an ordinary way a measurement setup drifts, usually across several years and several people.
A page view
A conversion tag placed on every page rather than only on the confirmation step, so arriving counts as converting.
A click on a phone number
Tapping a number is measurable; whether the call connected, or lasted more than two seconds, is not. Intent is counted as outcome.
The same enquiry, several times
A confirmation page that can be reloaded, or a visitor who submits twice, counted as two customers.
A form that was started, not finished
An event that fires on the first field interaction, or on a validation step, rather than on a successful submission.
An engagement signal imported from analytics
An analytics event built to measure interest, such as scroll depth or time on page, imported into the advertising account as a conversion.
A newsletter signup, or a brochure download
A genuine action, and a real conversion for some businesses. It becomes a measurement problem when it is counted alongside sales enquiries as though the two are worth the same.
Could someone be inflating our conversions deliberately?
It is possible, and it is not where the evidence usually points. Overcounting is far more often inherited, accidental, or the residue of a setup built for a different purpose, and no check can distinguish intent from incompetence in any case.
This deserves a direct answer rather than a diplomatic one, because it is what a reader in this situation is actually thinking. The reason to be sceptical of the deliberate explanation is that the accidental ones are so much more common: a tag on every page, an imported analytics event, a phone click counted as a call. Each of those produces inflated numbers with nobody deciding anything.
What a check can establish is what is actually being counted. That is the useful thing to know regardless of how it came about, and it is a considerably more productive conversation to have with an agency than an accusation would be.
What this cannot tell you Intent is not observable and this service does not speculate about it. A report will describe the mechanism, not the motive.
Which actions should count as a lead?
Actions your business would recognise as somebody genuinely making contact: a completed enquiry form, a connected phone call, a booked appointment, a submitted quote request. Not a page view, not a button click, not a partially completed form.
The practical test is whether a real person would follow it up. If nobody in your business would act on it, it is not a lead, and counting it as one puts an acquisition cost against something that will never become revenue.
Softer actions are not worthless and they are not the problem. The problem is counting them alongside sales enquiries as though the two are equivalent, because that is what teaches automated bidding to pursue the cheaper one.
Why does this get worse over time rather than staying steady?
Because automated bidding learns from whatever it is told is a success. If an easily triggered action counts as a conversion, the system finds more people who perform that action, which is not the same as more customers.
This is the mechanism that makes overcounting expensive rather than merely inaccurate. A reporting error you can correct at the end of the quarter costs you a bad decision. A bidding signal pointed at the wrong action costs you continuously, and it gets more efficient at doing so.
It also explains a pattern that confuses people: conversion volume climbing steadily while actual enquiries stay flat. The system is succeeding at the goal it was given.
What this cannot tell you Whether a specific conversion action is used as a bidding goal is a setting inside the account, and an external check cannot see it. This is the single most important thing to establish, and it takes about ten minutes with access.
Does a high conversion rate mean the campaigns are working?
Only if the conversions represent customers. A conversion rate is a ratio of clicks to counted actions, so if the counted action is trivial to trigger, a high rate is a measurement artefact rather than a commercial result.
An unusually high conversion rate for your sector is worth a second look for exactly this reason. It can be excellent performance. It can also mean the bar for what counts has been set very low.
Find out what your conversions are counting.
The free check observes what your site sends and when, including whether a conversion request fires before anybody has done anything.
Should newsletter signups and brochure downloads count as conversions?
They can, as long as they are not counted alongside sales enquiries as though the two are worth the same. The problem is rarely the action itself, it is the equivalence.
Two practical approaches work. Assign values that reflect what each action is actually worth to you, so a brochure download and a quote request are not treated as interchangeable. Or keep the softer actions out of the conversions the campaigns optimise toward, and track them separately as interest rather than outcome.
How do I compare Google Ads conversions against our own records?
Start with counts over a fixed period rather than trying to match individual records. Take a full month, count the enquiries your business actually received, and compare that to the conversions the advertising reported for the same month.
Expect a gap in both directions and pay attention to its shape. Conversions substantially above enquiries points at something being overcounted. Enquiries above conversions points at routes that are not being measured, most often phone calls. A stable ratio you can explain is a healthier sign than a perfect match, which usually means somebody has forced the two to agree.
If conversions have been counted wrongly, can the history be corrected?
Not retroactively in any meaningful sense. Past conversion data stays as it was recorded, and correcting the setup changes what happens from that point forward rather than what already happened.
This has a consequence worth planning for: after a correction, reported conversions often fall, sometimes sharply, and reported cost per conversion rises. Nothing has got worse. The number has started describing something real, and anybody comparing month against month without knowing that will reach the opposite conclusion.
Is a click on our phone number the same as a phone call?
No, and the difference is one of the more common sources of overcounting. A click on a number is a measurable intention. Whether the call connected, lasted more than a few seconds, or was somebody who dialled by accident, is not visible from the website at all.
This site is careful about the same distinction on its own pages. The phone number in the footer measures clicks on it, never calls, and it would be a poor look to blur a line here that the audit exists to point out elsewhere.