A difficult first week is a forecast of the whole engagement

The sample you already have

By the time a project starts you have had several interactions with the client: an enquiry, a conversation or two, a negotiation, some arrangement about starting. That is a small sample and it’s not a random one, because it happened during the period when both sides were being deliberately agreeable.

Which is what makes early friction so informative. Difficulty at the point where everyone is trying is not an unlucky week. It is close to the best behaviour available, and the rest of the engagement is unlikely to improve on it.

What the early signs actually are

Disregard for your time is the clearest. Meetings rearranged at short notice, calls that run far beyond what was arranged, questions sent at midnight with an expectation of an answer. None of these is outrageous in isolation and all of them describe how the following months will go.

Reluctance to agree anything in writing is the second. A client who deflects every attempt to define the work is either uncomfortable committing to a description or wants the freedom to redefine it later, and both are expensive.

Then there’s how they talk about their previous supplier. Occasionally a predecessor really was incompetent. When every past supplier was, the common element is not the suppliers, and the flattering position you currently occupy is the same position they occupied at the start.

And there is haggling that continues after the price has been agreed. A negotiation is normal. Reopening a settled number is a statement about how agreements are going to be treated.

Why it gets ignored

Because work is wanted, and because each individual signal has an innocent explanation. They were busy. It was a stressful week. Their last supplier may genuinely have been poor.

Each of those is plausible, which is exactly the difficulty — the signals are only convincing in aggregate, and by the time enough have accumulated the project has started. The people who avoid this aren’t better judges of character; they simply decided in advance which behaviours they treat as disqualifying, so the judgement is not being made in the moment.

What to do short of declining

Most of the time the answer is not refusal but structure. A written scope. A deposit. Staged payments. A stated change process. Defined channels and hours for contact rather than open availability.

These are all reasonable things to have as standard, which is what makes them usable — none of them requires you to explain that the client seems difficult. And they work, because a great deal of client behaviour is a response to whatever structure exists, and a supplier who has none is inviting the client to supply their own.

A price that reflects the anticipated difficulty is also legitimate. Work that will consume more attention costs more to deliver, and pricing it accordingly is straightforward rather than punitive.

Declining, when it comes to that

Sometimes the right answer is no, and the reasons that make it easy to say are the ordinary ones: capacity, timing, or that the work isn’t quite what you do. There is no obligation to explain a judgement about how somebody behaves, and doing so achieves nothing.

The cost of declining is one project. The cost of accepting badly is that project, plus the attention it takes from everything else, plus whatever work you could not do while it consumed the diary. That comparison is rarely close once it has been made explicitly.

Being wrong in both directions

This should be held loosely. People have bad fortnights, organisations have chaotic periods that pass, and a client who was disorganised at the start sometimes turns out to be excellent once the work is under way. Treating early friction as certainty would mean declining perfectly good clients on the strength of one rearranged meeting.

The useful way to hold it is as a probability rather than a verdict — a reason to add structure and to watch, not a reason to refuse. And it is worth remembering that the client is doing exactly the same assessment of you, on the same thin evidence, from the other side of the table.

The habit worth building is a short private note after the first fortnight of any new engagement, recording how it has gone so far. It takes two minutes and it captures an impression that will otherwise be overwritten by everything that follows. Read a dozen of those notes back against how the projects actually turned out and you will have something better than instinct: your own record of which early signals, in your own trade, actually predicted trouble.