Intake is the part of a professional services business that nobody owns and everybody touches. It grew rather than got designed. Somebody set up the enquiry form, somebody else added the spreadsheet, a calendar got involved, and at no point did anyone sit down and decide how it should work.
That is normal. It is also why intake is usually the highest-return thing to fix and the last thing anyone looks at.
Here are three signs, in rough order of how expensive they are. None of them requires an audit to spot. You will know within a minute whether each one applies.
None of these needs a bigger team. Each one is a system doing a job that a person is currently doing from memory.
Sign one: the same details get typed into more than one system
The test: follow one new client from first enquiry to first invoice, and count the number of places their name and contact details get entered by hand.
If the answer is more than one, you have this sign.
The cost that is easy to see is the minutes, and the minutes are not the problem. The problem is that the same fact now lives in several places and can differ in several places. A client updates their billing address. One system gets updated, because it was the one open at the time. The others quietly keep the old one.
Nothing raises an alarm, because from each system's point of view nothing went wrong. Each is internally consistent and confidently holds a different answer. You find out when an invoice goes to the wrong address, which is roughly six weeks after the moment you could have caught it.
What fixing it looks like: every piece of information gets entered once, into the one place that is allowed to be right. Everything else reads from there instead of keeping a copy. This is duller than it sounds and it is the foundation the other two fixes sit on.
Sign two: follow-ups happen only when someone remembers
The test: if the person who usually chases outstanding enquiries were off for two weeks, would anything chase them?
If the honest answer is no, you have this sign.
This one is worth taking seriously because of when it fails. A follow-up that depends on memory does not fail randomly. It fails on your busiest weeks, because that is when attention is scarcest. Which means it fails precisely when you have the most enquiries in flight and the most to lose from dropping one.
The failure is also invisible in a specific way: you never find out about the client you did not follow up with. They do not complain. They go elsewhere, and you record nothing, because from the inside a lost enquiry and an enquiry that was never serious look identical.
That is the worst kind of problem to have. It costs money and produces no evidence that it is costing money.
What fixing it looks like: the chase runs on a schedule whether or not anyone remembers. The person's judgment goes into what the message says and which client needs a different approach, not into remembering that today is day five.
Sign three: you find out about a problem after the client does
The test: think of the last thing that went wrong on a matter. Did you hear it from your own system, or from the client?
If it was the client, you have this sign. This is the expensive one.
The work required to fix the underlying problem is usually small. A date was missed, a document did not go out, a status was not updated. On its own, twenty minutes of work.
The work required to rebuild the client's confidence is not small, and it cannot be done in twenty minutes. Because what the client learned is not that one detail slipped. It is that your system did not know before they did, which raises a reasonable question about what else it does not know.
For a professional services firm, that is the whole product. Not the filing or the advice memo, the confidence that details are handled. This sign spends that confidence in one visible transaction rather than gradually.
What fixing it looks like: the system watches for the conditions that mean something has gone wrong, and tells you first. Not a dashboard someone has to remember to open. An alert that arrives when the thing happens.
The pattern underneath all three
Read the three fixes together and the same shape appears each time.
| The sign | What a person is doing from memory | What should do it instead |
|---|---|---|
| Details typed more than once | Keeping several copies in step | One record, everything else reads from it |
| Follow-ups depend on remembering | Tracking who is owed a chase and when | A schedule that runs regardless |
| You hear about problems last | Noticing that something looks off | A check that watches and raises the alarm |
In every row, a person is currently acting as the memory and the monitoring for a system that should be doing its own. That is the actual diagnosis. It is not that the team is careless or too small. It is that they have been handed a job that humans are structurally bad at, which is remembering to check the same thing reliably forever, on busy weeks included.
People are excellent at the judgment calls sitting on top of all three: which client needs a different approach, whether this exception is worth making, how to word the difficult message. Those are the parts worth protecting, and they are exactly what gets squeezed out when attention is spent on remembering.
What to take away
If one or more of these landed, three things to do before looking at any software:
- Trace one real client end to end. Not from memory, actually follow the path from enquiry to first invoice. The count of manual re-entries is usually higher than anyone in the firm would guess.
- Ask what breaks if one specific person is away for two weeks. Whatever the answer names is a process currently running on a single person's memory.
- Fix sign one first. Follow-ups and alerts both depend on trusting a record. Automating on top of records that disagree with each other sends wrong information faster, which is worse than doing nothing.
None of this needs a bigger team. It needs a system that runs the repetitive part and flags the rest for you. The AI does the busywork. You still make the calls.