Choosing the first customer segment

The earlier problem card said “night-shift workers.” That label still hides important differences. A nurse who cannot leave a ward, a guard at a remote gate, and a receptionist who can step outside for five minutes work at night, but they face different delivery constraints. A segment is a reachable group sharing a relevant use and purchase situation. It is not simply an age bracket or occupational label. Choosing the first segment makes a small experiment interpretable; it does not mean the venture will never serve others.

First separate three roles. The user consumes the meal. The buyer approves or makes the payment. The site authority permits the service to reach the workplace. At a public pickup point, one person may perform all three roles. In a hospital, the nurse may be the user, the nurse or employer may pay, and the facility decides whether a service can enter. A user's interest cannot substitute for site permission. Compare candidate segments on how often the occasion occurs, how consequential the present difficulty is, whether relevant people can be reached for research, and what constrains delivery and payment. These are operational criteria for choosing a test, not a universal formula for finding the best market.

Consider three fictional groups in the same area. Group A consists of 80 workers on night shifts in a health facility. The cafeteria closes for five hours, some workers cannot leave their stations, and a shift coordinator can connect the team with ten volunteers. Nobody has yet confirmed the site's pickup rules. Group B has 140 office staff buying lunch near three open food outlets; reaching them in a public area is easy. Group C has 35 workers at a remote industrial site with few apparent alternatives, but access requires permissions that have not been explored and there is no contact channel. These headcounts are group sizes, not forecast orders. Choosing B merely because it is largest could lead to testing a problem already well served. Choosing C merely because alternatives seem scarce could produce a study the team cannot conduct safely. A is a sensible first candidate for *research* because the occasion is specified and interviews are feasible, provided the team treats operational permission as a gate before any delivery pilot.

Make the reasoning inspectable with three columns: what is known, what is assumed, and what blocks the test. For A, the fictional cafeteria hours and contact route are known, while the number of people who actually need a meal, willingness to pay, and delivery permission remain unknown. For B, competing options are visible, but their adequacy for a particular subgroup is unknown. For C, remoteness and reported limited alternatives are known, but neither a usable research channel nor entry conditions are established. A weighted score with arbitrary numbers would disguise these differences. Instead, set minimum conditions for learning: voluntary contact with workers from more than one shift, no pressure to participate, and a statement from the authorized site person about where the team may operate. If one condition fails, revise the study or use another segment. Do not call the group “validated.”

A sharper segment description improves the interview. “Health facility workers” does not identify the closing-time occasion. “Workers on duty between midnight and 5 a.m. who cannot readily leave their stations” does. Avoid narrowing it after seeing friendly answers. Before recruiting, write inclusion criteria such as at least two relevant night shifts in the previous month and note exclusions. Do not define the segment as “people who want our meals”; that builds the desired answer into the sample. Include people whose current alternative works, because their evidence may disconfirm the venture premise.

An interview segment and a launch segment are different decisions. A facility might allow ten interviews yet prohibit food sales inside, or workers may show interest while the actual payer is an employer with a long purchasing process. This lesson therefore ends with a research selection memo, not a launch commitment. State the selected group, the reason it outranks alternatives, how people will be reached, the three roles, and the unknowns that govern the next decision. The U.S. Small Business Administration lists demand, location, market saturation, and price among market-research questions; it does not certify the fictional comparison above. If site access changes, the memo should make it easy to change the selection.