A research memo is only as useful as the brief behind it — here is what to put in yours so the answer settles a real decision.
A research memo is only as good as the brief behind it. Hand an agent a vague question — tell me about the market — and you will get a vague memo back: pages of general background, no recommendation, nothing you can act on. Hand it a sharp brief and the same agent returns something you can decide from. The difference is rarely the agent. It is the handful of things in the brief, and none of them takes long to write.
Write the decision in one sentence before you write anything else. For example: should I open a weekly home-cleaning service for busy families in Corvallis, Oregon, or stick to one-off deep cleans? That single sentence tells the agent what to compare, whom to study, and what a useful answer would settle. Without it, the agent fills the gap with background — market definitions, history, general trends — because you never told it what was at stake. If a finding would change your mind, name it: a cheaper rival, a licensing hurdle, a neighbourhood already well served. That is the test your memo must pass, and it is the brief Atlas, GIGSTERbot's research agent, works best from.
Next, list the constraints that rule options in or out. Budget, geography, timeline, and what you will not do are not background detail; they are filters. A memo that recommends a city-wide launch is useless if you can only serve two neighbourhoods this year. A pricing analysis that assumes a premium service misses the point if your whole plan is to undercut on price. State the budget range you are working within, the area you can actually serve, and the date the decision has to be made by. Constraints narrow the research to the world you live in, which is the only world the decision has to work in.
Vague audiences produce vague research. Busy families is a start; families with two working parents, a house over a certain size, and no cleaner yet is a brief an agent can work with. The same goes for competitors: name the three to five businesses you actually lose work to, or describe the type precisely if you do not know their names yet. Include what you already know about their prices and promises, and say plainly what you do not know. Guessing at a competitor's price and presenting it as fact helps no one; marking it as something to check gives the agent a question worth answering.
Ask for the shape of the answer, not just the subject. A useful research memo usually has the same spine: who the market is and how it breaks into segments, how competitors position themselves and what they charge, a clear-eyed view of strengths, weaknesses, opportunities and threats, and a recommendation with reasons. Say you want that structure, and say what the recommendation should cover: which option to choose, what to watch out for, and what to do first. If you need the memo to fit a particular use — a conversation with a partner, a page in a business plan — say that too, so the length and tone match the job.
The most valuable line in a research memo is often the one that says: this figure is uncertain, check it before you spend money. Ask for that explicitly. Tell the agent to separate solid facts from estimates, to mark its sources, and to list the numbers that matter most to your decision so you can verify them independently. A memo that dresses up a guess as a fact is worse than no memo at all, because it lets you act with false confidence. Atlas marks sources and flags anything unverifiable rather than dressing it up, but the habit is worth asking for in any brief: honesty about uncertainty is a feature, not a weakness.
When the memo lands, read the recommendation first, then the evidence behind it. Does the evidence actually support the recommendation, or does it point somewhere else? Are the load-bearing numbers — the ones your decision hinges on — marked as solid or uncertain? If something is unclear, ask a follow-up question while the brief is still fresh; a short clarification is cheaper than a second full memo. Once the decision is made, the opening the research found becomes the core of everything else: Lyra can turn it into launch copy that names the customer's problem and your answer to it, rather than describing your business in general terms.
If you are launching rather than just deciding, you do not have to commission the research on its own. The Launch Pack bundles the research memo with launch copy and a logo starter from a single questionnaire, so the brief you write once feeds all three pieces — and they all point at the same opening.