Introduction
Most Pipedrive accounts are never audited. They are configured once during onboarding, then adapted in small increments — a stage added here, a custom field there, a rule someone needed for one quarter — until the setup describes a sales process nobody actually runs any more.
This is a checklist you can work through in about an hour, in order. Each check tells you what to look at, what a bad result looks like, and the fix. Open Pipedrive in one window and this list in another; you will not need anything else.
Pipeline and Stage Structure (about 15 minutes)
Check 1: Count your stages
Look at: the stages in each pipeline.
Bad result: more than seven stages, or any stage whose exit criterion you cannot state in one sentence. If two people on the team would move the same deal to different stages, the stage is not defined.
Fix: write the exit criterion for each stage as something checkable — “next meeting booked and budget confirmed” rather than “customer is interested.” Merge or delete any stage that cannot carry one.
Check 2: Look for the graveyard stage
Look at: deal counts per stage.
Bad result: one stage holding a disproportionate share of open deals, usually somewhere in the middle. That is where deals go to be neither advanced nor closed.
Fix: the pile-up is a symptom, not the disease — it means the exit criterion for the next stage is unclear or the deals are dead and nobody has closed them. Work through the oldest ten and force a decision on each.
Check 3: Multiple pipelines
Look at: how many pipelines exist and who uses them.
Bad result: pipelines created for a campaign or a person and never retired. Deals sitting in a pipeline nobody reviews are invisible to your forecast.
Fix: archive unused pipelines. Keep separate pipelines only where the sales process genuinely differs, not where the segment differs.
Activity and Deal Management (about 20 minutes)
Check 4: Open deals with no scheduled activity
Look at: a filter of open deals where the next activity is empty.
Bad result: anything above roughly a fifth of your open pipeline. These deals have no mechanism for progressing — nobody has committed to doing anything about them.
Fix: this is the single highest-value check in the audit. Work the list, and either schedule a real next step or close the deal. Then keep the filter as a saved view and open it weekly.
Check 5: Overdue activities
Look at: activities scheduled in the past that are still open.
Bad result: a long tail of overdue items, especially any more than two weeks old. Once the overdue list is long enough to feel hopeless, the team stops looking at it, and the activity system stops working entirely.
Fix: declare an amnesty. Clear the backlog in one sitting — complete, reschedule or delete every item — so the list becomes credible again. A trustworthy short list beats an accurate long one nobody opens.
Check 6: Rotting settings
Look at: whether deal rotting is configured per stage, and what the thresholds are.
Bad result: rotting switched off entirely, or a single threshold applied to every stage. Early-stage deals and late-stage deals go stale on very different clocks.
Fix: set a threshold per stage, based on how long deals in that stage normally take when they are healthy. Then actually review what the indicator flags, or you have installed a warning light nobody looks at.
Check 7: Deals past their expected close date
Look at: open deals whose close date is in the past.
Bad result: more than a handful. Every one of them is either a date nobody updated or a deal nobody wants to close as lost, and both corrupt the forecast.
Fix: for each, set a date the rep will defend or close it lost with a reason. Do not let dates slide silently — a close date that has moved three times is telling you something the pipeline number is not.
Data Quality Checks (about 15 minutes)
Check 8: Required fields at stage transitions
Look at: which fields are required, and where they are enforced.
Bad result: either no required fields at all, or a long list enforced on every edit. The second is worse — it makes routine updates painful, and reps route around pain.
Fix: enforce a small set at the stage transitions where the information actually becomes knowable. Nothing should be required before the point in the process where a rep could reasonably know it.
Check 9: Duplicate organizations and people
Look at: your organizations and contacts, sorted alphabetically, and Pipedrive’s duplicate detection.
Bad result: the same company under two spellings, each carrying part of the history. Duplicates split the record of the relationship and cause two reps to call the same account.
Fix: merge them. Then look at how they got there — usually an import or a form that creates records without matching — and fix the source, or you will do this again next quarter.
Check 10: Won and lost reasons
Look at: your closed deals from the last quarter and whether a reason is recorded.
Bad result: reasons missing on most closed deals, or a free-text field with fifty variations of the same three answers. Either way the loss teaches you nothing.
Fix: use a short structured list — budget, timing, competitor, no decision, product fit, champion left — plus one optional free-text field for context. Six options people actually use beat twenty they pick at random.
Check 11: Custom fields nobody fills
Look at: your custom fields and how often each is populated.
Bad result: fields populated on a small minority of deals. A field filled a third of the time cannot be reported on and is pure friction for the reps who do fill it.
Fix: delete or archive them. If a field matters, make it required at the right transition; if it does not matter enough to require, it does not matter enough to keep.
User Permissions Review (about 10 minutes)
Check 12: Active users and leavers
Look at: your user list and last login dates.
Bad result: accounts belonging to people who have left, or users who have not logged in for months. Both are a security exposure and, in the second case, a sign that someone’s deals are unmanaged.
Fix: deactivate leavers and reassign their open deals explicitly rather than leaving them owned by a dormant account.
Check 13: Who can see and change what
Look at: permission sets and visibility settings.
Bad result: everyone an administrator. It means anyone can change stage definitions, delete fields or bulk-edit deals, and configuration drift becomes untraceable.
Fix: restrict administrator rights to the people who own the configuration. Reps do not need to be able to redefine the pipeline in order to sell.
What to Do With the Findings
You will finish the hour with a list of a dozen or so items. Do not try to fix them in one go. Sort them by whether they change a decision this quarter: deals with no next step and overdue activities come first, because they represent revenue in motion; duplicate records and unused custom fields can wait a week.
The harder part is that this audit is a snapshot, and the setup drifts again from the day after. Doing it continuously rather than once is the point at which it stops being a chore — SalesBond analyses how the pipeline and CRM are structured, including stage logic, required and missing fields, data quality, duplicates and next-step rules, and turns what it finds into a prioritised Fix Plan. Approved fixes can be applied automatically where the connected CRM allows it, and nothing changes without an authorised user approving it.
FAQ
How often should I audit my Pipedrive?
The full pass is worth running quarterly, and after any change to your sales process or team structure. The two activity checks — deals with no next step, and overdue activities — are worth a weekly glance.
What is the single most valuable check on this list?
Open deals with no scheduled next activity. It is fast, it needs no configuration, and it consistently surfaces more at-risk revenue than any other single filter.
Should I delete old deals or leave them?
Close them as lost with a reason rather than deleting them. Deleting destroys the history that tells you why deals in that segment do not close; closing them keeps the lesson and cleans the pipeline.
Who should run the audit?
Whoever owns the configuration — usually a sales operations lead or the sales manager. It is worth doing alongside one rep, who will explain in five minutes which fields are actually painful.
How long until the fixes show up in the numbers?
Data quality improves immediately. Forecast accuracy takes a full sales cycle to reflect the change, because the pipeline has to turn over before the cleaner inputs work through.
Conclusion
An hour and thirteen checks will tell you more about the state of your sales process than a month of pipeline reviews. Stage structure shows whether the process is defined, activity and rotting checks show whether it is being worked, data quality checks show whether you can trust what you measure, and permissions show who can quietly change it.
Block the hour this week and start with check 4 — open deals with no next activity. It takes two minutes to build the filter, and whatever it returns is the most useful list in your CRM today.
