Most RevOps "alignment" projects are tooling projects in disguise. New routing rules. A cleaner scoring model. A tighter Salesforce report. The work happens at the system layer because that is where it is easiest to ship — a routing change has a date, a pull request, a release note, a slide for the next QBR. The system layer is legible.

I have been brought in for a lot of these projects. The output is usually good. Routing improves, scoring recalibrates, the lifecycle program stops double-counting opportunities. And then about a quarter later, the same RevOps lead is on the phone asking why marketing and sales are arguing again, and whether maybe a different attribution model would help.

It would not. The problem was never attribution.

The standard alignment audit

The conventional RevOps alignment project starts with a tool inventory. What systems do we have. What are they doing. Where are the gaps. A spreadsheet emerges, severities are assigned, work is sequenced. The first ninety days fix the visible problems — the SLA on lead routing, the broken Marketo sync, the report that has been wrong since Q2.

This is real work and it should be done. Cleaner tools make every downstream decision faster. But the audit treats the stack as if it were the alignment, and the stack is not the alignment. The alignment is upstream of the tools, in a layer most audits never touch.

What that audit misses

The layer that actually matters is the one where marketing, sales, and finance disagree without realizing it. The team agrees on a metric in a meeting. The metric gets shipped into Marketo or Salesforce. Six months later, the original meaning has drifted, and three different people are pulling reports from three different definitions of the same number, each one technically defensible.

A small example. A team I worked with had agreed in Q1 that an MQL meant a contact with a score above sixty plus a "Marketing-Qualified" stage transition. Q2, a new VP joined and shifted the team's focus toward intent data — accounts hitting a 6sense surge score. The lifecycle program was never updated, but the QBR slides started reporting "qualified accounts" as the primary metric. By Q3, the SDR team was working two different lists derived from two different definitions, and the VPs of marketing and sales had each built reports defending their own number. The system had not broken. The agreement had drifted, and the system was still enforcing the old one.

Cleaning up Marketo would not have fixed this. The agreement needed to be re-formed first.

A more useful lens

These days, when I scope a RevOps engagement, I spend the first week not in the tools but in conversation. I ask marketing, sales, and the SDR team the same five questions separately, and I write down where the answers diverge. The questions are operating ones.

What counts as a qualified opportunity this quarter, and who decides. When marketing hands an account to sales, what is sales actually supposed to do in the first forty-eight hours. When sales rejects an account, what should happen to that person — back to nurture, hard suppression, recycled in ninety days, or quietly disappeared. When an opportunity stalls, who notices and what fires. How does the team agree that a quarter went well.

The answers almost never match. The list of disagreements is the actual roadmap. The Marketo and Salesforce work is downstream of resolving those disagreements — once the team agrees on what each transition means and who owns it, the tooling changes are obvious. Without that agreement, the tooling changes are guesses dressed up as best practices, and the next vendor inherits whatever assumptions you wrote into the system.

How this changes the recommendation

If you score a RevOps engagement on this dimension before you start, the scope changes. A team strong on operating agreement but weak on system enforcement has a tooling problem — they need clean execution against decisions they have already made. The audit works. A team weak on operating agreement but strong on system hygiene has the opposite problem. Their stack is fine. What they need is a facilitated week where the leaders agree on what they were actually trying to measure, and the system gets rebuilt to enforce that agreement honestly.

The most expensive mistake I see is teams running the tooling audit when the operating layer is the real issue. The tooling audit produces visible progress, leadership feels good for a quarter, and the underlying disagreement quietly compounds. By the time someone notices that the new routing rules are routing leads into a definition nobody believes anymore, the audit budget is spent, the trust between marketing and sales has eroded, and the next RevOps engagement inherits a more complicated problem than the one it sold against.

The diagnostic

If your team cannot answer five specific questions about a closed-lost deal from last quarter — why that account moved when it did, what marketing did, what sales did, who decided to disqualify, and what the system said about the decision — the issue is not the stack. No CRM, no MAP, and no scoring model substitutes for an agreement nobody has written down.

A clean Salesforce instance under a confused operating model produces faster, more confident bad decisions. The cleanup that matters is the conversation.

If you want more on how I scope and run RevOps engagements, full notes are on the owned page: vadim-koenen.github.io/revops.