Hygiene campaigns fail because they are framed as tidying. Reframed as the input layer of the forecast — with the cost of each defect made visible — the same work gets done and stays done.
Every revenue organisation has run a pipeline hygiene campaign. Most have run several, because the first ones did not hold. The pattern is familiar: a push before quarter end, a spike of updated records, a gradual decay back to baseline over six to eight weeks.
The campaigns fail for a reason that has nothing to do with discipline. They are framed as tidying — as making the CRM look right — and tidying is exactly the kind of work that loses to a live deal every single time. Nobody has ever been fired for a stale close date.
Reframe: hygiene is the forecast's input layer
The forecast is a function of the pipeline. If the pipeline contains deals in the wrong stage, with dates nobody believes, owned by people who left, then the forecast is a function of noise and no amount of modelling sophistication downstream will fix it.
That reframing changes what you ask for. Instead of 'please update your deals', you ask for the specific facts the forecast depends on — and you can say what each one is worth.
Name defects, not fields
'Incomplete records' is not a defect anyone can act on. These are:
- A deal in Commit with no identified economic buyer.
- A close date inside the quarter that has already been pushed twice.
- An opportunity whose last activity predates the segment's median gap for deals that closed.
- A stage that behaviour contradicts — Negotiation with no pricing discussion in any thread.
- An amount that has not changed since creation on a deal that has been through discovery.
- An owner who no longer works here.
Each of those is checkable, each maps to a specific forecast error, and each can be shown to the person who can fix it with the evidence already attached.
Ask for a fact the forecast depends on and you get an update. Ask someone to tidy their pipeline and you get a sigh.
Every hygiene programme that stuck
Attach the cost
A hygiene rule that cannot be traced to a consequence will be treated as bureaucracy, correctly. Each rule should carry its own justification, computed from your own history: deals that entered Commit without an identified economic buyer closed at some materially lower rate than those that did, and that difference, multiplied by the deals currently in that state, is a number.
Publish those numbers next to the rules. The argument stops being about process compliance and becomes about a quantified risk to a number everyone already cares about.
Detect at the moment, not in the cleanup
A monthly hygiene report is a list of defects that have already done their damage — they have been in the forecast for up to four weeks. The same rules, evaluated continuously, surface a defect while the deal is still live and the correction still changes something.
This is the difference between hygiene as a periodic campaign and hygiene as a property of the system. The rules are identical. The latency is everything.
Automate the correction where the answer is knowable
A large share of hygiene defects have an answer sitting in a system you already own. The economic buyer's title is in the email thread. The pricing discussion is in a call transcript. The competitor mention is in a support ticket. The right stage is inferable from the deal's own behaviour.
Where the answer is knowable, asking a human to type it in is a waste of the one resource the quarter is short of. Propose the correction with its evidence and let the owner accept it. Where the answer is genuinely unknown — is this deal really going to close this quarter — no system should pretend otherwise, and the ask should go to the human.
What good looks like after two quarters
Hygiene stops appearing on anyone's roadmap. There is no campaign, because defects are surfaced and mostly resolved within a day of appearing. The forecast review no longer starts with an argument about whether the data is right. And the pipeline number in the board deck is the same number the system holds, without a reconciliation step in between.
That is an unglamorous outcome, which is precisely why it is worth building toward. The organisations that get there stop spending management attention on whether they can trust their own records, and spend it on the deals instead.