Impact mapping for product managers

Impact mapping is a planning technique I use to connect a product goal to the people, behavior, and work that might help us reach it. Instead of starting with a list of features, I start by asking what outcome matters, whose behavior needs to change, and what we could do to support that change. The map gives the team a shared line of reasoning without pretending that the plan is a guarantee.

I find it useful when a roadmap has become a collection of requests. A requested feature can be reasonable and still be several steps removed from the result the organization needs. Impact mapping helps me expose that distance so I can test whether the proposed work is a credible way to create value.

Start with the goal

I put one specific goal at the top of the map. The goal might be to help new teams reach a first successful workflow, reduce avoidable support demand, or increase repeated use of a valuable capability. I write it as an outcome rather than a deliverable. “Launch a guided setup” is work; “help administrators complete setup with less uncertainty” is a goal worth investigating.

I also define the time frame, audience, and signal I will use to judge progress. I do not need a perfect metric before beginning, but I do need enough precision to notice when the map is drifting. If the goal is broad enough to include several unrelated problems, I split it before mapping.

The goal is a hypothesis about value, not a command to make a number rise at any cost. I keep guardrails visible when they matter, such as preserving reliability, user trust, or the quality of an important workflow.

Identify the actors

Next I ask who could influence the goal. An actor is not always a generic “user.” It may be a new administrator, a team lead approving a workflow, a buyer who needs confidence before adoption, or an internal support specialist helping someone recover from a failed step. Different actors can face different barriers, so I avoid putting them in one audience bucket too quickly.

I write down what each actor is trying to accomplish in the context of the goal. I also note whether the person directly experiences the problem, makes a decision, or influences someone else’s behavior. This distinction changes the product question. A manager’s approval behavior is not the same as an end user’s day-to-day usage.

I treat actors as a prompt for investigation, not as proof that I understand them. Interviews, support conversations, product data, and observation can challenge my first map. When evidence shows that the real audience differs from my assumption, I update the map rather than forcing the evidence to fit.

Map the intended impacts

For each actor, I write the behavior or impact that would contribute to the goal. I prefer verbs that describe a meaningful change: adopt the workflow, invite a teammate, return to complete a task, choose a safer configuration, or recognize that a problem has been resolved. I ask why that change would matter and what would make it plausible.

This is where I separate an outcome from a proxy. A click on a new button may be useful evidence, but it is not automatically the impact I want. If I want teams to complete a workflow successfully, a click can be an early signal while completion quality or repeated success may be closer to the real outcome. I write both the intended behavior and the evidence boundary so the team does not confuse activity with value.

I also look for competing explanations. If usage rises after a release, the change may reflect a seasonal pattern, a policy change, or a forced migration rather than an improvement in the product. An impact map helps me form the question; it does not remove the need for careful analysis.

Connect impacts to deliverables

Only after I have written the goal, actors, and intended impacts do I discuss deliverables. A deliverable is something the team could create, change, test, or remove. It might be a clearer explanation, a workflow change, an integration, a training aid, or a small experiment. I keep several options under an impact so that the map does not quietly turn one solution into a requirement.

For every proposed deliverable, I ask which impact it is meant to influence and what assumption connects the two. If I cannot explain that connection, I either remove the item or move it into a question for discovery. This simple check has helped me avoid accepting a feature because it sounds familiar or because a stakeholder has already invested in the idea.

I do not treat every branch as equally important. I mark high-risk assumptions, especially where the proposed work depends on a behavior we have not observed or a constraint we have not verified. A small test can be more useful than expanding the map with more speculative features.

Use the map in conversations

I bring an impact map into product, design, engineering, and stakeholder discussions as a reasoning artifact. I ask where people agree, where the evidence is thin, and which branch would change our next decision. The map gives disagreement a location. Someone may support the goal but question the actor, agree about the actor but doubt the behavior, or accept the impact but prefer a different deliverable.

I keep the map lightweight enough to revise. It is not a contract, a project plan, or a substitute for a delivery backlog. When learning changes the problem, I update the branch and record what changed. When work ships, I connect the result to the relevant impact and review whether the expected behavior actually moved.

A practical starting exercise

I take one roadmap item and rewrite it as a goal. Then I name the primary actor, describe the behavior that would contribute to the goal, list two or three possible deliverables, and write the riskiest assumption under each connection. I choose the assumption most likely to change the decision and plan the smallest credible test.

Impact mapping works for me when it keeps product work attached to a reason. It lets me discuss outcomes without hiding the practical work required to pursue them, and it gives the team a way to learn before a long feature list becomes a commitment.

Next step

Build stronger product strategy, prioritization, and outcome-focused decision-making skills in the Product Manager Certification. Subscribe to the Product HQ newsletter for practical frameworks and career-ready product lessons.

Kevin Lee
Kevin Lee
Kevin is a Co-Founder of ProductHQ. He has worked as a VC at Pear Ventures where he invested in and partnered with early-stage founders on product & growth to help them build the foundations of category-defining companies. He has worked as a Product Manager at AltSchool (backed by Andreessen Horowitz, Founders Fund, First Round Capital, Mark Zuckerberg, John Doerr and other exceptional investors). Previously, he was a Senior Product Manager at Kabam (acquired by NetMarble and Fox for a combined $1bn+), where he worked on products through all lifecycles in San Francisco, Vancouver, and Beijing and helped grow one of the company’s products to become the third largest revenue generating product in the company portfolio. In a former life, he worked in Technology Investment Banking at Merrill Lynch. He is also the author / co-author on 10+ gaming patents.