Someone fills in a form on your website. To marketing, that is a conversion: the lead counts in the monthly report. Sales opens the same lead in the CRM, sees a student who wanted a white paper, and skips it. At the end of the month, marketing reports ten new leads and sales reports three usable contacts. Both figures are correct. They just count different things.
That is the core of almost every gap between sales and marketing figures. It is rarely a data error and rarely bad faith. The two teams count different things, at different moments, and attribute the same result to a different cause. Until that is agreed, the figures keep drifting apart, however well the systems are connected.
Marketing counts behaviour, sales counts opportunities
Marketing measures what people do: view a page, fill in a form, open an email. Each of those signals can produce a lead, and a lead that gives off enough signals becomes a marketing qualified lead (MQL), ready to hand over to sales. Sales judges the same person on something else: is there budget, does this person have a say in the decision, is there a concrete need with a timeline. Only then does it become a sales qualified lead (SQL).
The handover from MQL to SQL is where the figures part ways. Marketing measures broadly and inclusively, sales narrowly and selectively. That is how it should be: marketing fills the funnel, sales filters it. It only becomes a problem when each team uses its own count to claim success or to blame the other. Marketing feels sales does not follow up enough, sales feels marketing delivers too little quality, and both have the figures to prove it.
The same result, a different cause
Even when both teams agree on what a lead is, their figures still differ through attribution: who gets the credit for a deal. Marketing often looks at the first interaction, or spreads the credit across every touchpoint in the customer journey. Sales looks at the last conversation before the signature. Neither is wrong. But when marketing reports on a multi-touch model and sales on the last contact, they tell a different story about how the same deal came about.
The same applies to the question where a customer actually came from. An advert, a conversation at a trade fair and a referral may all have contributed to the same deal. Which of the three you count is a choice, and it is a choice you make together.
The moment of measurement shifts the figures
The third cause is timing. Marketing measures at the moment something happens; sales only counts once someone has picked up the lead. A lead that arrives on 31 December belongs to the old year for marketing and to the new year for sales. For a single lead that makes no difference. Across a month boundary, with dozens of leads and a follow-up time of a few days, such shifts produce differences that nobody can explain without laying the two registrations side by side.
When the difference becomes a problem
The question is not whether the figures are identical. They never will be, and they do not need to be. The question is whether they point in the same direction. If both teams see a similar trend with slightly different absolute numbers, the difference is explainable and harmless. If marketing says it was a strong quarter and sales says it disappointed, something is wrong with the agreements.
A second signal is how meetings go. If every discussion starts with which number is right instead of with the decision on the table, the gap has grown too large. That is a different issue from figures that contradict each other because the same fact is recorded differently in two systems; we wrote about that in the risks of deciding on inconsistent data. Here the data is right. The agreement about it is missing.
First the definition, then the integration
Because the difference lies in the definitions, that is where the solution starts. Three agreements on paper are enough to remove most of the discussion:
- What a lead is and when it becomes an MQL: which signals count, and which do not.
- When a lead passes to sales, and what sales does with a lead that does not meet the bar: send it back with a reason, not leave it lying.
- How a deal is attributed, with one model that both teams use in their reporting.
Then comes the technology, and it matters more than it seems. As long as marketing works in a marketing platform and sales in a CRM that does not know the other system, every lead exists twice, with its own status in each system. An integration between the two makes the agreement enforceable: a lead is created once, a status change in one system appears immediately in the other, and both teams report from the same source. The argument over whose list is right disappears by itself, because there is only one list left. Why the same lead grows apart in two places is explained in our article on duplicate customer profiles.
A shared report closes it off: one overview with the same definitions for both teams, so nobody reasons from a private export anymore. Plan a fixed moment to test the definitions against practice. Markets change, and last year’s MQL is not always this year’s.
Do you recognise the recurring discussion between sales and marketing? Start with the three agreements, not with the software. Once the definition is clear and you want your CRM and marketing platform to enforce it, that is work for API integrations: a lead that exists once and has the same status everywhere. We are happy to think along with you about which integration that takes.