Four Lines and a Date

What a research agency should receive from you, and what usually stops them from getting it.

It usually starts in November. Somebody asks how much we should spend on market research next year. You give a number. Everyone forgets about it until March.

Then in March somebody needs consumer research. A campaign is coming, or a product, or something else. You sit down with the business owner and start writing the brief.

Then the brief gets circulated. Every stakeholder who reads it has another ten things they need to find out about the consumer. All ten have to fit into this particular project. A few weeks go by consolidating opinions.

Then, on a Friday afternoon: did you send the brief? We need to start as soon as possible.

By then the brief asks about everything except the decision that started it.

Here is what makes that expensive. The brief is the only document that crosses out of the building. Whatever the agency understands about your decision, they understand from that page.

They were not in the meeting. They did not hear what the owner said in the corridor. They have no idea which option got ruled out in February. They get the page.

So write the page for someone who was not in the room. A brief is a description of a decision you have not made yet. That settles what belongs on it.

Four lines and a date.

First line. The decision you have to make. Not the topic. Not the market. The one that is still genuinely open. Two decisions means two briefs.

Second line. What you will do depending on the answer. Every option you would actually take, and the result that sends you to each one. That tells the agency what has to be measured well and what can be measured roughly. If you would do the same thing whatever the study comes back with, do not buy the research.

Third line. The one thing you must know, and what you are deliberately not asking. If price cannot move this year, say so, and stop paying people to ask about price. Every question you cut is budget that goes into the one that matters.

Fourth line. Who has to accept the result, and what they run. Not who signs the invoice. Sales, production and the owner do not accept the same evidence. Tell the agency which of them is waiting and they will build the case for that person. Leave it out and you get work that is correct and gets picked apart on method.

Finally, the date. Put the day the decision gets made on it. Agencies work backwards from deadlines, and that date tells them what is worth proposing and what is fantasy. Put a second one on it too. A session before the deck gets designed, when the top lines go in front of exactly those people.

Filled in, it takes a paragraph:

The decision is whether to launch the new range nationally in September or hold it to one region for six months. It has to be made on 15 June. If the people currently buying the cheaper brand see a reason to switch, it goes national as planned. If they just see another version of what they already buy, it stays regional and the campaign money goes to holding on to the customers we have. What we must know is which of those two it is. We are not asking about price, packaging or the name. All three are fixed for this launch. Sales and the owner have to accept the answer. Both of them believe it is a reason to switch. Neither has spoken to one of those buyers this year. Top lines go in front of both of them on 28 May.

Notice how much is not on that page.

The questions you want asked, for a start. Method is the agency’s job, so let them propose it and defend it.

Here is another job that stays with you. Turning a business problem into a research problem takes knowing what is being decided, what is on the table, and what the company will not do whatever comes back. The agency cannot see any of it. Leave it to them and they work backwards from your questions to a decision that may not be yours. It is tempting to leave it to them, because they are the experts. They are, but in how to find out. You are the only one who knows what has to be decided.

There is also a limit on how much of the internal situation goes out with it. Send the functions, what each of them believes today, and what each would refuse to accept. Keep your read on the politics to yourself. Who is protecting a budget, who lost this argument last year. None of it improves the research, and it gives the agency room to manage your stakeholders instead of answering your question.

That is the brief. Almost nobody sends it, and the reasons are always the same three.

The first is timing. By the time the brief goes out, your date is fixed and the plan is already moving. The slot is booked. The launch has a window. Whatever comes back cannot change much of it. That is how a study turns into a receipt, and nobody ever decides to make it one.

The second is money. Something always gets cut.

If the brief names the decision, cutting is easy. You keep whatever separates the options and drop the rest. If it does not, nothing in the scope matters more than anything else, so you cut a bit off everything. Smaller sample, fewer locations, less depth, and nothing left that decides anything.

A full budget does not save you either. It buys a bigger sample, a second wave, a segmentation nobody asked for. Not clarity. Ninety slides, all correct, none of them deciding anything. Nobody is being cynical. An agency cannot invent your decision, so they sell you rigour instead.

The third is alignment, and it does the most damage. You are not writing this alone. Several people stand behind you and they do not want the same outcome, so the brief gets negotiated. What lands on the page is the version everyone could sign, phrased broadly enough that no possible answer embarrasses anyone who has to approve it.

The money and the veto also sit in different places. The person releasing the budget gets consulted on scope. The person who can reject the conclusion often finds out what was asked when the results are presented.

Which is why disagreement about the conclusion arrives dressed as disagreement about method. The sample was too small. The wording was leading. This was fielded in the wrong month. All of it arguable. None of it the real objection. All of it unanswerable once fieldwork has closed. The study gets relativised rather than rejected, and the plan proceeds as it would have anyway.

None of these three constraints is going away. So work inside them.

Write the decision before you circulate anything. Ask people to review a brief and they add questions. Show them a decision and they argue about the decision, which is the argument worth having.

Write the options next. If they all lead to the same place, keep the money. When the options do not split cleanly, agree the number that decides it. Twenty percent say they would switch and we launch, ten and we do not. Agree it before fieldwork, because afterwards it gets argued against whatever came back.

Write the exclusions before the budget conversation, not after it. When the money shrinks, the list of what you are not asking tells you where the scope comes off. Written afterwards, it just describes what you already lost.

Get the veto onto the page early. The beliefs of the people who can reject the result, in their own words, contradicting each other. Circulate it and let silence stand as agreement. The argument happens at the brief or at the debrief. At the brief it costs an hour. At the debrief it costs the study.

Put the interim date in the brief itself. Left as an informal courtesy, it is the first thing dropped when the timeline slips. Objections raised before the deck is designed can still be answered, sometimes with data already collected. Three weeks later the only answer is to run it again.

A good agency does half of this for you. They send the decision back in their own words at proposal stage, and they ask who has to agree with it for the work to be used. Let them.

Sometimes the study was never meant to settle anything. Research gets bought to postpone an argument. Everyone agrees to wait for the data, the room relaxes, and the fight reconvenes in three months unchanged. Nothing on the page fixes that, because the brief did exactly what it was for.

Four lines and a date takes about fifteen minutes, which is roughly what the calendar allows.

1. The decision you have to make.
2. What you will do depending on the answer. Every option, and the result that sends you to each.
3. The one thing you must know, and what you are deliberately not asking.
4. Who has to accept the result, and what they run.

And the date. The day the decision gets made, and a session before it where the top lines go in front of those people.

Nobody at the agency can write those five things for you. Send the page without them and they will run it anyway.