Seven CRM Reports a Founder Should Actually Read

Enquiries by source

Count enquiries by where they came from, week on week, with spend attached where there is any. What you are looking for is a change in shape rather than a change in total. A steady total that has quietly shifted from one source to another is a more important finding than a small dip, because it usually means something changed in a channel and nobody noticed.

Pipeline by stage with age

Count and value by stage, with average age in stage displayed alongside. This is the closest thing a small business has to an early warning system. Read it as a shape: a thin qualified stage means a revenue problem in a few weeks, a fat proposal stage means a follow-up problem right now.

First response time by channel

Median time from customer contact to first human reply, split by channel and by working hours. This is the report with the best ratio of effort to improvement, because the fix is almost always ownership and routing rather than money.

Open records with no next action

A count, not a chart. It measures the share of your pipeline that nobody is actively working. Most businesses running this for the first time find the number uncomfortable, and the discomfort is the point: those records are either work that needs an owner or deals that should be closed.

Win and loss reasons

Closed deals grouped by reason, with values. This only works if reasons are captured at the moment of closing from a short forced list. If most of your losses are recorded as other or price, look at the list rather than the market: price is the default answer people give when the real reason is harder to write down.

Conversion by stage

The percentage of deals moving from each stage to the next, over the last few months. This tells you where the process leaks rather than where deals sit. Combined with time in stage from the weekly report, it usually points at one specific step that deserves attention, and fixing one step is a manageable project in a way that improving conversion generally is not.

Revenue by customer and concentration

Revenue per customer over the last twelve months, sorted, with the share taken by your largest few shown explicitly. Founders often discover that a larger share of revenue than they assumed comes from a handful of accounts. That single number should influence how much effort goes into retention against acquisition, and it should probably influence how nervous you are.

Alongside the seven, keep two live counts visible rather than reported. The first is unanswered enquiries older than a set interval, which is a queue and needs acting on within the hour rather than reviewing on a Monday. The second is deals whose stage time limit has been breached, which is a prompt for a decision rather than a measurement of the past.

The distinction between a report and a live count is worth holding on to. Reports describe what happened and inform decisions about next week. Live counts describe what is happening and demand action now. Teams that put live counts into weekly reports end up reviewing problems that could have been solved on the day they appeared, and teams that put reports on a live screen create noise that everybody learns to ignore.

Three habits do most of the work. Put the weekly four on one page and look at them at the same time every week, ideally immediately before the pipeline review so the numbers inform the conversation. Write the decision next to each report rather than only the number. And delete anything that has not changed a decision in two months.

One more, which is less obvious: put your data quality numbers on the same page. Next-action coverage, unowned records, suspected duplicates. When those sit beside the sales numbers, the relationship between them becomes visible, and the argument about whether the reports can be trusted gets settled with evidence instead of opinion.

Related reading: what a CRM is, reporting features, lead management software, CRM for small business, sales automation, and book a walkthrough.