Most buying guides for marketing automation software compare the same things: how many contacts a tier allows, how many journeys you can build, whether it sends text messages, what the template editor feels like. All of that is real, and none of it decides whether the software will ever tell your CFO which campaign produced which deal.
That depends on three fields and what happens to them between the first click and the closed deal. You can test all three in a demo, in about twenty minutes, before you sign anything.
Marketing automation software lives at the front of the funnel. It sees the form fill, the email click and the page visit. The sale lives in the CRM, at the other end, and the thing that connects them is a handful of fields that either travel with the person or do not.
The usual failure is not a missing integration. Most tools sync with most CRMs. The failure is that the sync overwrites. A lead arrives from a webinar link, and the campaign field says so. Two weeks later the same person clicks a retargeting email, and the field now says that. The sale closes in March, and the report credits the last thing the person touched before the sales call, which is usually a nurture email, because that is where almost everyone ends up.
Nobody decided that. It is a default, and it is the reason a pipeline report from the tool and a pipeline report from the CRM disagree for the same quarter. The tool credits what it saw most recently; the CRM credits what it was told when the deal was created. Both are internally consistent. Neither is the number the CFO wants, which is the campaign that created the opportunity.
The first-touch campaign, frozen at creation. When a person first becomes a record, the campaign that brought them is written once into a field that nothing else is allowed to overwrite. It is not the "source" field the tool updates on every visit. It is a second field that is set on creation and then read-only for everything except a named administrator.
The same campaign, copied to the deal when it is created. When the lead converts or a person creates a deal for a contact, the frozen campaign is copied onto the deal, as a value, at that moment. It is not looked up later through the contact, because the contact can change, merge or be deleted. The deal carries its own copy.
A cost key the tool can read back. The campaign identifier must be the same string the finance system uses on its bill lines, so that cost can be joined to pipeline without a spreadsheet in the middle. The tool does not need to hold the cost. It needs to carry the key and let a report group by it.
We set out the thinking behind the first two in the one table a CFO will trust: attribution is a bookkeeping problem, fields on the lead and a status a person sets. This piece is the buying version. Which tool can actually do that bookkeeping without a developer?
Ask for a demo environment connected to a sandbox CRM, or ask the vendor to show the following on their own. If they cannot, that is information.
Step one, create and freeze. Create a lead from a form on a page that carries a campaign identifier in its link. Look at the lead in the CRM. Find the field that holds the campaign. Now ask: can this field be locked from the tool's own updates? Can it be set only when empty?
Step two, disturb it. Have the same person click a link from a different campaign, then submit a second form. Look at the lead again. The frozen field should still show the first campaign. The ordinary "latest source" field may show the second. Both should exist, and you should be able to tell them apart in a report.
Step three, convert and compare. Convert the lead to a contact with a deal. Look at the deal. The first-touch campaign should be on the deal, as a field of the deal, without anyone typing it. Then edit the contact's campaign field and look at the deal again. It should not change.
Step four, group by it. Build a report of deals by first-touch campaign, with the deal amount. If the tool can do this natively, good. If it needs the CRM's reporting for that step, also fine, as long as the field is available there and not trapped in the tool.
Each failure tells you something specific. If step one fails, there is nowhere to put a value that survives. If step two fails, the report will credit the last touch whatever you intend. If step three fails, the deal and the marketing record will drift apart the first time anyone cleans a contact. If step four fails, the numbers exist but nobody can read them without an export.
A team that buys on features and tests this after go-live finds out in the first quarterly review, when the pipeline report shows nurture email creating most of the deals. The honest response is to rebuild the fields, which means migrating history: the original campaign for existing contacts is in an old export, or in nobody's memory. Fixing it forward takes a week. Fixing it backward is a guess, and the CFO knows it is a guess. It is the same gap we described in the report that hides the number the CFO wants, only this time the report is the tool's own.
The cheaper moment is before the contract. A vendor that passes the four steps has handled the part of the problem that decides whether the report is trusted. A vendor that fails them may still be the right choice for sending email, but you will be buying a sender, and the measurement will live in the CRM and a spreadsheet, which is a decision you can make with open eyes.
There is one more question to ask, and it is about people rather than software. Who owns the list of campaign identifiers? If nobody does, the field fills with variants: "webinar-oct", "Webinar October", "webinar_oct_2026". The test passes in the demo and fails in production, because the field is frozen and wrong.
The owner is a person, not a tool. They issue a campaign identifier before a campaign gets a budget, and the same identifier goes on the links, the ad account and the bill line. That one habit is worth more to the report than any feature on the pricing page.
In discovery we run the four steps against the tool you already have or the one you are considering, in your CRM, and show you which fields survive and which do not. Discovery is no-risk: you pay only if you go ahead. The work to set the fields, the copy rule and the identifier list carries a guaranteed estimate, so if we estimate low, we absorb the difference.
The test is simple enough to run yourself, and we encourage it. What we add is the accountability for it still passing six months later, after the tool has had two updates and someone has merged three hundred contacts.
Watch the 6-minute guided demo, then book a walkthrough: we read your CRM, ad accounts and content and show what a governed system would do differently. You pay only if you proceed. Or see how we approach it.
More on the same problem:
"Something isn't working." "We're outgrowing our tools." "We're afraid to make the wrong move." No-Risk Discovery is a short, practical conversation that gets you clarity before you commit to anything big. We'll tell you if we're a fit. If we're not, we'll tell you that too.
Say who you are and ask. A person replies within one business day.
