How to work out what to fix first in your business

This is a solvable problem, and solving it does not require a consultant. What follows is roughly the process we run, written so you can do it yourself.

Start with the path a customer actually takes

Not the one on the process document. The real one. Pick a recent enquiry and follow it end to end: where it arrived, who saw it first, how long it sat, what got typed in and where, who chased it, and how long from first contact to money in the bank.

Do this for three or four real jobs — one that went well, one that went badly, one that was lost. The gap between them is usually where your answer is. People are often surprised by how much of the delay is waiting rather than working: a quote that takes twenty minutes to write can still take four days to send.

Put a number against each delay

This is the step almost everyone skips, and it is the one that turns a list into an order. You do not need precision. You need enough to compare.

• How many enquiries do you get a month, and what share never get a proper follow-up?
• What is an average job worth, and what share of enquiries do you win?
• How many hours a week does your team spend on admin that a system could do?
• What does an hour of that person’s time cost you, including everything?

Multiply them out. Enquiries that go cold, valued at your existing close rate, is one number. Admin hours times fifty-two times the hourly cost is another. Both are usually larger than people expect, and — this matters — both are calculated entirely from figures you already have. No industry benchmarks, no assumed improvement.

Two rules for this to be worth anything. Use your own close rate, not a better one you hope to reach: the point is what the current situation costs, not what a perfect one would earn. And do not include anything you cannot evidence — a number you had to guess at will be the number someone argues with, and the whole exercise dies there.

Sort by cost, then by what has to happen first

Now you have a list with numbers on it, order becomes obvious in a way it was not before. Two adjustments to a naive "most expensive first" ranking:

• Some fixes are prerequisites. Sending more traffic to a site that cannot take a booking makes the leak bigger, not smaller. Plumbing before pressure.
• Some fixes are cheap and immediate. A single automated follow-up on unanswered enquiries can be live in a week. Doing one of those first funds patience for the longer work.

Write down what you are not doing, and why

The value of this exercise is as much in what it rules out. Six months later, when somebody asks why you did not rebuild the website, the answer should exist in writing: because quoting was costing four times as much and the site was not the constraint.

Without that, every abandoned idea comes back every quarter, and you re-litigate the same decision with worse recall each time.

When this is worth paying somebody for

Do it yourself if you have the time and can be honest about your own operation. Both are harder than they sound — the second more than the first. People who run a business are, reasonably, attached to the way they run it, and the steps that exist because "that is how we have always done it" are exactly the ones that become invisible.

The other reason to bring somebody in is that the output has to survive a room. If this has to get past a business partner, a board or a lender, a document with your own numbers in it and an outsider’s name on it carries differently from the same argument made by the person who wants the budget.