CRM

Availableapp.erp.io/crm/analytics

Analytics

Three numbers matter: where leads come from, how fast they move, and where they stop. Everything else is decoration.

Attribution
First touch, from ingest
Velocity
Days per stage
Conversion
Stage to stage
Export
CSV

Source attribution

Every contact carries the source recorded at ingest and never overwritten. That makes source attribution genuinely reliable here in a way it usually is not — it is a fact stamped at creation rather than a reconstruction from later activity.

app.erp.io/crm/analytics
Analytics

Last 90 days

SourcesVelocityConversion
Leads by source — bar length is volume, figure is win rate
Portal signup11%
Website form6%
Phony call24%
Referral38%
Outbound3%
Source performance. Volume and conversion side by side, because the biggest source is rarely the best one.

The shape in that figure is the common one and it is worth naming: the sources that produce the most leads are usually not the sources that produce the most customers. Judging a channel on volume is how a business ends up doing more of the thing that does not work.

Velocity

Days spent in each stage, by opportunity and on average. A stage that everything sits in is the most actionable finding in the module — it means either the exit criteria are unclear or the next action belongs to somebody who is not doing it.

PatternUsually means
One stage far slower than the restUnclear exit criteria, or a dependency outside the team.
Fast movement then a cliff at one stageThe stage before it is being entered too early.
Everything fast, low win rateQualification is too loose upstream.
Slow everywhereToo many stages, or nobody updating them.

What to ignore

  • Open rates. Mail privacy features fire tracking pixels on messages nobody read. The number is not measuring what it appears to measure.
  • Activity counts. Calls made and emails sent measure effort, not progress. A team optimised on activity produces activity.
  • Pipeline value alone. Without stage weighting and velocity it is a number that only ever goes up.

What this does not do

No multi-touch attribution

First touch only. There is no model distributing credit across touches.

No revenue attribution

Analytics stops at the opportunity. Actual revenue lives in Accounting.

No custom dashboards

The views are fixed. Export to CSV for anything else.

No cohort analysis

There is no retention or cohort view.

Questions

Can we see performance per person?

Yes, filtered by owner.

Does analytics include lost reasons?

Yes, if they are recorded on the lost stage.

Can we export the raw data?

Yes, CSV, or through the API.