The three levers that actually matter in SMEs
Capacity (how long work sits in a queue and how long it takes to do), rework (errors, disputes, and round-trips that burn the same people twice), and speed to cash (time from work done to money in the account). You do not need a perfect data warehouse to compare projects — you need a one-page story the board and the shop floor both recognise.
Book a free systems review if you want a second pair of eyes on the numbers in your own context. Our offers cover audits, build, and ongoing system care once the change is in production.
How to get numbers you will not abandon
Perfection is the enemy. Pick one or two high-volume paths — e.g. quote to job, or invoice to payment — and for one week, note:
- Rough hours spent in handoffs (who waits on whom, and for how long).
- A count of rework events in plain language: wrong line on a quote, missed PO, re-keyed data, dispute that bounces a week.
- Days from completion or invoice to cleared payment (even if you use "calendar days in the last month" as a proxy).
If you are shy on history, use estimates the team can defend in a room, not a decimal from nowhere. Ranges are fine; the point is a comparable score across ideas — not a forecast to the penny.
| Question | What it tells you |
|---|---|
| If we do nothing for six more months, what is the repeat cost? | Momentum to fix vs defer |
| Who is the human bottleneck, and is it a training fix or a system fix? | Avoids buying software for a one-hour training win |
| Is the main pain in in, through, or out? | Focuses the build or process change on the right part of the pipe |
Time, rework, and speed to cash — in one breath
Time is not just clock time — it is attention. If a process waits two days in an inbox, that is cost even if the actual work is 20 minutes.
Rework is often the hidden line item: a wrong cut on a quote, a credit note, a return visit, a call back. Capture it as count × typical minutes or pounds, not a vague bad feeling.
Speed to cash is where leadership often sees the case first. Shorter time from done to paid often lines up with clearer status, fewer disputes, and fewer manual steps. Even an approximate day-count is enough to rank one change against another.
Comparing projects on one page
Try a one-page view you would put in front of a non-technical owner (no Gantt, no 40-slide deck):
- Idea in one line (what work stops being awful).
- Which lever — capacity, rework, cash, or a mix — and a gut band of benefit in the next six months (low / med / high is enough to start).
- What has to be true in week one (training, a single new habit, a cut-over date).
- What we are not doing in this bet (scope kill).
Then sort by clarity and repeat cost, not who shouted loudest in the last meeting. The goal is a decision that can survive a busy Tuesday — not a model the accountant is alone defending.
Where we help: we work with Teesside and the North East and remote partners. A process audit or build is easier to justify when the why is visible on one page — book a systems review and we can help you get there without boiling the ocean.