Market research

Published: 2026-09-12 · Last reviewed: 2026-09-12 · Reading edit: 2026-09-12

Someone asks how big the market is. You find three reports with three different answers, but none tells you whether the companies you can reach would actually buy. Start with the decision in front of you. This guide helps you gather enough useful evidence to choose where to test, without turning research into an endless slide deck.

Start with a decision

Use this guide before committing a quarter of product or marketing work to a new audience. Write the choice and deadline first: “Should we test inventory reconciliation with regional distributors this quarter?” Record what evidence would make you defer the test. If no finding could change the decision, label the exercise background reading rather than validation.

Build an evidence trail

1. Choose which companies you are researching. Specify geography, operating model, company size, buyer role and the task being purchased. Count buying organizations rather than every employee who might use the product. Distinguish subsidiaries from independent purchasing units.
2. Describe the current workflow. Identify tools, services, manual work, approval paths and existing spending. Read procurement documents and actual job descriptions alongside vendor material. Vendor category definitions may exaggerate the audience relevant to your offer.
3. Separate facts from estimates. For each number record its source, publication date, population and definition. A global software forecast cannot establish demand in your reachable segment. Show the assumptions connecting a broad count to a serviceable market.
4. Investigate recent behavior. Speak with people who recently bought, delayed, rejected or worked around a solution. Ask what happened and what they tried. Do not turn a favorable response to your pitch into a completed purchase decision.
5. Look for contradictions. Compare the strongest buying case with a company that appears similar but did not buy. Record whether the difference is urgency, authority, integration, economics or simply missing evidence.
6. Decide what to try next. Produce a short decision memo with the strongest finding, most important unanswered question, what could go wrong, and who will run the next test.

Worked example

A fictional team identifies 900 distributors in a region. Only 180 appear to run multiple warehouses; 70 have the supported accounting system. Neither filter proves a budget exists. Six interviews reveal that month-end reconciliation matters, but four firms outsource it cheaply. The team tests the two in-house workflows before estimating demand from all 70 organizations.

The useful finding is that some otherwise suitable companies already solve the problem cheaply. That changes whom the team should speak to next. Six selected interviews cannot tell them what proportion of the whole market will buy.

Their short decision note could look like this:

Question	What the fictional team writes
What are we deciding?	Whether to test a reconciliation workflow with regional distributors
What did we learn?	Four of the six interviewed firms outsource the task cheaply
What is still unclear?	How often the two in-house teams struggle and what the problem costs them
What happens next?	Ask those teams to walk through a recent month-end reconciliation
What would make us pause?	The existing process works well enough that changing it has little value

Copyable research brief

Decision and deadline:
Buying organization and exclusion rules:
Current task, alternative and spending:
Evidence: claim / source / date / population / limitation
Conflicting observation:
Reachable count and assumptions:
Smallest next test, owner and cost ceiling:
Proceed / narrow / defer, with reason:

Review the result

Ask a colleague to trace your three most important conclusions back to evidence. A colleague should be able to understand a number’s population and date without needing you to explain the slide. Refresh material assumptions before spending, especially where regulations, budgets or technology requirements may have changed. Stop adding sources when the next decision is clear; investigate the remaining uncertainty through a test.

Try it with your own work

Choose one market you are considering. Write the decision you need to make, then list three facts that could change your mind. Find evidence for the most consequential one first.

Sources and scope

• US Small Business Administration: market research and competitive analysis distinguishes direct research from existing data and includes indirect competitors. Its business-planning guidance is a starting point, not a B2B market-size estimate.

The example is fictional; any numbers illustrate the method rather than a benchmark. Adapt the worksheet to your own situation.

What to read next

Use segmentation to choose groups that require different actions, then idea validation to test the proposed offer.

Chapter guide · All playbooks

---

Copyright © 2026 Ivan Xu. All rights reserved. See the copyright and reuse terms.

Canonical source: github.com/weilun88313/B2B-Playbook