You have been asked whether a small product should add an English edition. You collect competitor pages, messages from prospective users, and a rough translation estimate. By the end of the afternoon there are plenty of links, but no clear recommendation.
This walkthrough turns that collection into a one-page decision brief. The product, evidence, and findings below are fictional examples; they are not market research or TodoNote's own launch results. Use the structure with sources you have actually checked.
Give the research a decision to support
“Research international expansion” leaves almost everything open. A bounded question makes it easier to decide which sources matter and when you have enough to write.
Decide whether to test an English edition next month
Reader: The product owner who will decide what to build next.
Deliverable: A one-page recommendation comparing a limited test with translating the whole product.
Constraints: A small team and no dedicated English support role.
Questions: What evidence of demand exists? What must a first-time user understand? Who would maintain translations and answer support questions?
Finish line: A recommendation with sources, open assumptions, and a next step that fits the team's capacity.
Separate the evidence you need
Group the research by questions rather than by the order in which browser tabs were opened. In this case, use three sections: demand, first-use experience, and ongoing cost. Each source should help answer one of them.
A competitor's English website shows what that competitor publishes. It does not prove your users want an English edition. A request from a prospective user is more direct, but it still does not tell you how many people would activate or pay. Write down what each source can and cannot support.
Attach a source and a limit to each claim
A request for an English setup guide
Source: A dated message from a prospective user; keep a link to the original in the working note.
Observation: The person could not follow the current setup instructions and asked for an English guide.
Interpretation: Language may be preventing this person from trying the product.
Limit: One request does not establish wider demand or willingness to pay.
Next check: Offer a short English guide and ask which steps remain unclear.
Record the date you checked a changing source, such as a pricing page. Keep excerpts short and explain why they matter. When a number is an estimate, label it as an estimate and note what it includes.
In TodoNote, keep these entries in the research task's note. Link to original materials where they already live. A source that contradicts your preferred answer deserves a place in the note too.
Make uncertainty visible before drafting
Suppose the illustrative evidence shows interest in English setup, but the support workload remains unknown. Create a follow-up task: “Walk through first-time setup in English and list the questions it creates.” Its result will help estimate work that a word count alone would miss.
Another useful task is “Check who can maintain the English guide after the next release.” A language edition creates ongoing work, so the initial translation estimate cannot answer the whole question.
Begin the draft while these checks are open. Mark the gaps explicitly. If a missing answer could reverse the recommendation, make the decision conditional or wait for that answer. If it only affects a later detail, explain the limit and continue.
Test the English setup journey before a full translation
Recommendation: Start with a limited English setup guide and a small, explicitly recruited trial.
Basis: The example request identifies a concrete setup barrier. The team's capacity makes a full translation difficult to maintain.
Alternative considered: Translate the whole product immediately. This would cover more screens but commits maintenance effort before demand is understood.
Uncertainty: We do not yet know whether people who finish setup continue using the product, or what support they need.
Next step: For this example, recruit five volunteers for one week. Observe whether each can create a first task without help; record the questions and support minutes.
Revisit when: The trial ends, or translation and support work exceeds the agreed capacity. Link each factual claim to its evidence entry in the working version.
Check the reasoning before polishing the prose
- Can the reader identify the recommendation without reading the whole research trail?
- Does each factual claim have a source, with estimates and assumptions clearly marked?
- Have you included a credible alternative and the evidence that could change your mind?
- Does the next step produce information needed for the next decision?
Finish when the brief is delivered and the required review is complete. Keep the final recommendation near the evidence and leave unresolved follow-ups visible. A week later, someone should be able to challenge the conclusion without reconstructing the entire research process.
If the work is interrupted before delivery, use a short current-state and next-step note to make it easy to resume.