"Pipeline hygiene" is almost always used to mean "a tidy CRM": fields filled in, consistent stages, no duplicates. That is data cleanliness, and it is fine, but it does not decide anything.
The definition that is useful is a different one: pipeline hygiene is how closely what your CRM says about a client matches what has actually happened with that client.
Put that way it stops being a maintenance chore and becomes a measurement. And a measurement can be calculated.
Why it matters more than tidiness
An immaculately tidy CRM can be completely wrong.
Every field filled in, every stage consistent, and at the same time three deals that have been dead for a month because the client stopped answering, and two more that look stalled because the rep has been writing from their phone and none of that gets logged anywhere.
That CRM passes any data-quality review and produces a false forecast. Which is why the measure is not "how many fields are filled in", it is "how much reality is missing".
The four indicators
1. Expired records
What percentage of your open deals has a close date in the past.
The cheapest indicator, and the most honest one about the state of the typing. An expired date does not mean the deal is going badly: it means nobody has gone in to correct it, and therefore that nobody is looking at that record.
Calculated entirely inside the CRM.
2. Time in stage
How many days each open deal has spent in the stage it is in.
Do not confuse this with "no activity for X days", which is what every CRM shows by default. They are not the same:
- "No activity for 40 days" means nobody has logged anything for 40 days.
- "40 days in Negotiation" means the deal has been where it is for 40 days.
The second does not depend on somebody remembering to write a note. HubSpot stores the date a deal entered each stage on its own, so this indicator also comes out of the CRM.
3. Next-step coverage
What percentage of your open deals has a task or a meeting with a future date.
A deal with no agreed next step is not advancing: it is waiting for somebody to remember it. It is the indicator that best predicts what will look identical a month from now.
Also from the CRM.
4. Activity leakage
What percentage of the real conversation with your clients never reached the CRM.
This is the one the CRM cannot calculate, and the one that decides.
You measure it by cross-checking the headers of your team's email and calendar against what is logged on each record, and keeping the difference. We call it the Activity Leakage Rate, and it is the only one of the four that has to look outside.
Why your CRM cannot give you the fourth
This is not a product failing, nor something missing in HubSpot or Salesforce. It is a limitation of construction.
A CRM logs activity from the contacts that exist in it. So it cannot show you activity from the ones that do not exist, nor activity that happened on a channel it is not connected to. It has nothing to compare itself against.
And it is not a rare case. We measured a freshly created HubSpot portal and out of the box, with no Sales extension installed, not one email was logged. Incoming mail from contacts that already exist is captured; outgoing needs an extension per person and per browser; nothing sent from a phone is logged; and nothing at all from somebody who is not a contact.
How to start measuring it this week
The first three are reports you can build in an afternoon with what your CRM already ships. For the detail of how to read them, see how to tell whether your sales forecast is reliable.
The fourth can be estimated by hand from a sample: take ten clients, count the emails in your inbox with each of them and the ones on their record. The difference over the total is your rate, with the precision a sample of ten gives you.
If you would rather have the number for your whole company than a sample, that is what CERUNA does: it reads the headers of your team's email and calendar, cross-checks them against the CRM, and names the difference. Start with a Revenue Blind Spot Audit: €750, report in 72 hours.
What discipline will not fix
The first three indicators improve with process and reminders, at least for a few weeks.
The fourth does not. It depends on every person manually copying into the CRM what they already did somewhere else, and that has not worked anywhere for twenty years. The way out is not getting them to log things. It is no longer depending on them logging things.