Notes for owners · Sales management
How to read a diagnostic report and decide what to do first
A way to sort the findings in a sales diagnostic or audit report by effort and by orders at stake, and to decide which fixes your own team can do and which need outside help.
The GullySales team · Updated 17 Oct 2026 · 5 min read
On this page
Read a diagnostic report in this order: first the findings with orders attached, then the effort each fix needs, then who in your business can do it. Put every finding in one of four boxes (quick and big, slow and big, quick and small, slow and small), start with quick and big, and treat the slow-and-small box as a list to leave alone. It is tempting to start with whatever is on page one, which need not be the finding worth most.
Start with the evidence, not the score
A score of 2 out of 5 on follow-up tells you nothing you can act on until you know what sits behind it. Before you read the plan, find the records each finding came from: which quotes, which months, which salesperson, which source.
Two things to look for. First, does the finding name a number of enquiries or orders, not just an adjective? "Slow response" is a mood. "Fourteen of thirty IndiaMART enquiries waited a day" (a made-up line, but the right shape) is a finding. Second, did anyone check it with the person who entered the data? A gap that looks like negligence is often a phone that was out of coverage.
If a finding has no records beside it, mark it as an opinion. It may still be right, but treat it as a question for your team.
Sort every finding into two boxes
For each finding, write two things next to it.
Orders at stake. Roughly how many enquiries or what order value sits behind this leak? Use the report's own numbers. If it gives enquiries and not rupees, multiply by your usual order size and treat the result as a rough figure, not a promise.
Effort to fix. Be honest in three levels: a habit (someone does something differently from tomorrow), a tool (a sheet, a CRM field, a template), or a rebuild (a change in people, process or system).
Then place each finding in the grid.
| Small effort (a habit or a template) | Large effort (a tool or a rebuild) | |
|---|---|---|
| Many orders at stake | Do first | Plan next, with a named owner |
| Few orders at stake | Do if it takes an afternoon | Leave it on the list |
Anything in the top-left box is where you start. Anything bottom-right can wait, even if the report scored it badly. A poor score on something that touches few orders is a tidy-up, not a priority.
Decide who fixes each one
A report that says "implement a CRM" has told you very little. Ask of each finding: can a person on my team do this with the time and skill they have?
| Kind of fix | Who usually does it | Example |
|---|---|---|
| A habit | You or your team lead | Reply to every new enquiry before the day ends |
| A document | Whoever sends quotes | An itemised quote page with a validity date |
| A record | The person who answers calls | One shared sheet with source, date and next step |
| A decision | You, the owner | Who gets which leads, and what discount is allowed |
| A build | Outside help, or a hire | Moving three years of data into a CRM |
Few of these are builds. If your report does not say which items your own people can do, ask it to. The paid diagnostic is one source of such a report.
For example: an auto-parts trader in Ludhiana
Take an invented trader in Ludhiana who sells brake components to garages and small fleets, with two telephone sellers and an owner who quotes the larger orders. The report has nine findings.
Three stood out once the owner sorted them. Quotes from the owner were chased by nobody (many orders, small effort: a date written on each quote). Discounts were different for the same garage depending on who answered (many orders, a decision only the owner could make). The CRM export was missing the source on most rows (few orders, large effort: a rebuild).
The mistake was reading page one. The report opened with the missing CRM source, because that finding was the easiest to prove, and the owner started planning a data clean-up that would have eaten weeks for little gain. Sorting the grid moved attention to the first two. The unchased quotes get a call date each, and a discount rule goes on one sheet of paper stuck by the phone, which annoys one garage that had been getting a deeper discount. The CRM clean-up stays on the list.
What to do with the leftovers
Do not delete the findings you are leaving. Move them to a short dated list titled "not now", with the reason beside each. Two things happen: the list stops nagging, and later, when conditions change (a new hire, a bigger season), you see which ones just became cheap.
Also be wary of a plan where every item needs outside help, or where every item is rated urgent. Real plans have a mix.
One page to take away
Before the meeting ends, write the plan on one sheet. It needs five columns: the finding, the box it sits in, the person who owns it, the starting number and the date you will count again. For the Ludhiana trader it read: "quotes chased: 0 of 14 last month; owner of the fix: the owner; recount at the end of next month." The starting number is what lets you say later whether the fix worked. Count again over a similar stretch of records, not a quiet fortnight.
When you disagree with a finding
Some findings will feel wrong. Ask for the sample behind it and read ten rows. If the rows prove it, accept it. If the rows are mis-entered or from a closed product line, say so and ask for the finding to be corrected or dropped. A report you cannot argue with is a report you will not use.
Take the report you hold, or the last one you were sent, and sort three findings into the grid tonight, starting with the one that has the most orders behind it. If the report is thin, or you have none and want a revenue growth audit, the FAQ covers how we begin, and you can tell us what you need.
Related services
