Why this matters
Reporting is why the CRM exists. Reps could track their own deals in a notebook. The account earns its licence fee when it answers management questions without a meeting: how much pipeline covers next quarter, where deals die, and whether last month's push worked. Insights can answer all of that natively, no export, no spreadsheet.
It only works downstream of discipline. Every report is a lens over fields, stages and activities. A conversion report over a pipeline with parking stages measures parking, not conversion. This is why reporting comes late in the six step method: it is the reward for the earlier steps, not a substitute for them.
The design decisions that matter
Decision one: start from questions, not from charts. Write down the five questions your team actually argues about, then build one report per question. Typical five: do we have enough pipeline, where do deals stall, what did we win and why, who needs coaching on what, and is the forecast moving.
Decision two: one dashboard per audience. The team sees the operating view in the Monday meeting. The manager keeps a coaching view grouped by rep. Leadership gets won revenue, forecast and coverage. Mixing audiences produces dashboards where everyone reads two tiles and ignores nine.
Decision three: group by fields you trust. Grouping won deals by industry only works if industry is a filled single option field, which loops back to field discipline. Every dashboard you want defines a field you must maintain.
Decision four: set two or three goals, not fifteen. Pipedrive goals track targets like won revenue per quarter per team. A small set keeps them meaningful. A large set turns steering into scorekeeping.
A worked example
The Monday dashboard we deploy for a ten-rep team. Six tiles, in reading order, each answering one standing question.
| Tile | Insights report | Question it answers |
|---|---|---|
| Pipeline coverage | Open deal value by expected close month | Is there enough for next quarter? |
| Funnel | Deal conversion between stages, trailing 90 days | Where do deals die? |
| Won this month | Won revenue against the monthly goal | Are we on target? |
| Stalled deals | Open deals with no activity in 14 days | What needs a push today? |
| Deal velocity | Deal duration by stage | Is the cycle getting longer? |
| Loss reasons | Lost deals grouped by lost reason, this quarter | What is beating us? |
The meeting runs top-left to bottom-right and takes twenty minutes. Nobody presents. The dashboard presents, people decide. The stalled deals tile alone changed behaviour in week one, because appearing on it twice in a row gets noticed.
Common mistakes
Measuring everything is first. Forty reports across nine dashboards means nobody knows which number is the number. Curate hard. Every tile you remove makes the remaining ones louder.
Second, activity leaderboards without context. Ranking reps on call volume produces calls, not revenue. Pair every activity metric with an outcome metric, and use the pairing for coaching conversations rather than public rankings.
Third, trusting reports over dirty data. A forecast built on deals with default close dates is fiction with an axis. The hygiene routine is not separate from reporting. It is the maintenance contract your dashboards run on, and the same dependency holds for the forecast view.
Maintenance
Quarterly, prune. Open each dashboard with its audience and ask which tile they used last month. Remove the silent ones. Check every report's filters after any pipeline or stage change, because renamed stages and archived pipelines quietly bend old reports.
Monthly, spot-check one number the hard way. Pick a tile, pull the underlying deal list, and verify ten records by hand. Ten minutes, and it is how you catch the filter that silently excluded a pipeline three weeks ago, before the quarterly review does it for you in front of the board.