Understand the audience by listening, and segment by need

Describe the audience by its situation

Target audience analysis is the second phase of the doctrine cycle and the one civilian campaigns most often skip. Its questions are plain. What conditions shape the current behaviour: cost, time, distance, habit, rules, past experience? What does the audience believe about the issue and about you? What are they doing now instead? Could they do what you ask if they wanted to? Doctrine also uses the word vulnerabilities for the motives and needs a message can address. In this manual read that word as needs: what the audience lacks that you can honestly supply. It never means a weakness to press on.

Narrow the audience until those questions have one answer. The general public has no shared conditions and cannot be analysed. A group defined by its situation can be: parents of children who missed a scheduled visit, or tenants in buildings affected by a proposed rule.

Research by listening

Find out by asking, in the open. Hold listening sessions hosted by people the audience already knows. Run a short survey with neutral wording. Have conversations in the style of deep canvassing, where the canvasser asks about the person's own experience, listens without arguing, and exchanges stories instead of delivering a script. Field experiments by Broockman and Kalla found that conversations of this kind can shift attitudes for months, and they are also the best audience research available to you. In every case tell people who you are, who is paying, what the answers are for, and that they may decline.

Expect to be surprised. Teams tend to assume the audience disagrees with them when the more common obstacle is practical. Ask what would make the behaviour easier before you ask what people believe. Record objections in the audience's own words, including objections to you.

Segment by need, never by weakness

Tailoring works: messages that address a group's actual barriers are more relevant and more likely to be read (Kreuter and Wray). Segment by barrier and by need, for example people who intend to act but have not, people with unanswered questions, and people who have decided against. Do not select anyone because they are easier to move while grieving, lonely, in debt, cognitively impaired, acutely stressed, or under age. If a list or an advertising platform offers you those categories, decline them, and screen your own lists so that people in those situations are not reached by accident with pressure they cannot weigh.

Decide in advance how you will treat people who have considered the question and said no. They are owed accurate information and an open door, not escalating contact.

Map trust and access

For each segment write down who they already trust on this subject, because stage four depends on it, and how they can be reached in practice: language, literacy, disability, working hours, internet access, and the channels they actually use. An audience you cannot reach in a form they can use is not yet your audience.

Collect less

Gather the minimum. Use aggregate figures where they will do. Do not buy or infer sensitive traits such as health conditions, immigration status, religion, or financial distress. Keep research notes separate from contact lists, set a deletion date, and check which privacy rules apply to any records you hold, especially health and education records. Profiling individuals to find the argument each one is least able to resist is a covert practice even when the message itself is true. It is described on the defensive side of this manual and has no place in this one.