The most expensive mistakes we are asked to fix were not technical failures. Somebody chose a system from a feature comparison, without anyone having watched how the work was actually done, and the team quietly went back to WhatsApp and a spreadsheet.
So we spend the first part of every engagement not talking about technology at all. Here is roughly what that involves.
1. Follow one customer, end to end
We take a real recent customer and trace everything that happened: how they found the business, what they saw, how they made contact, who responded, what was recorded, what was promised, how it was delivered, and whether anyone spoke to them afterwards.
This one exercise finds more than any questionnaire, because it exposes the steps that live in someone's head rather than in a process document.
2. Find the informal systems
Every business runs partly on things nobody would list as systems. A WhatsApp group. A notebook by the till. A spreadsheet one person maintains and everyone depends on. A senior member of staff who simply remembers.
These are not failures to be replaced. They exist because they work better than whatever official option was available. Software that ignores them gets abandoned. Software designed knowing about them tends to survive.
If your team invented a workaround, they were solving a real problem. Understand it before removing it.
3. Mark every manual handover
We note every point where a human moves information between two systems. Each one is a candidate for integration, and each one is a place where things currently go wrong. This list is almost always longer than the owner expects.
4. Ask what "a good week" means
Different businesses are optimising for genuinely different things. A hotel wants occupancy at the right rate. A salon wants repeat visits. A professional services firm wants a small number of qualified conversations. A retailer wants stock accuracy and margin after fees.
If we do not know which one you are, we will recommend the wrong priorities perfectly competently.
5. Only then, decide
With that map, the priority list is usually obvious and frequently different from what the client expected. Often the highest-value first step costs very little — routing enquiries into one place, correcting business information everywhere, connecting two systems that already exist.
Sometimes we conclude that the thing you asked for is not worth doing yet, and we say so. That conversation is uncomfortable and it is the most useful thing we offer.