Mapping alternatives and competition

When a founder says, “No competitor sells meals at 2 a.m.,” the comparison may be limited to an identical product. A worker can still bring food from home, buy something before the shift, wait until morning, use an existing delivery app, or skip a meal at that hour. An alternative is what a person does if the new offer is unavailable, whether a company sells it or the person produces it. Absence of an identical subscription therefore does not prove absence of competition. The specialist compares ways to accomplish the task and the limits of each, rather than collecting rival logos.

Build an alternatives matrix around one occasion: a worker on duty from 10 p.m. to 6 a.m. wants a meal at 2 a.m. and can leave the station for no more than two minutes. Useful columns include availability at that hour, pickup effort, the person's cost, preparation effort, reliability, and who controls access to the site. Do not fill an unknown cell with an estimate disguised as a fact. If delivery fees or building rules have not been checked, write “unknown” and plan how to find out. The weights of these columns may differ by segment. A worker with storage space may tolerate preparing food ahead of time, while one without it may not. The matrix is a device for comparing hypotheses and designing interviews, not a final league table.

In a fictional case, the team records four options. Home-prepared food is available if brought to work and appears inexpensive per meal to the worker, but needs preparation and storage. The cafeteria sells a meal for 9 fictional currency units at 10 p.m. and closes before midnight. A delivery app lists a meal for 15 plus a fee of 4. Its driver reaches the building gate, not the ward, and pickup may take ten minutes away from the station. The proposed service would charge 16 and use an internal pickup point, but neither permission for that point nor the operating cost is established. Comparing 15 plus 4 with 16 does not prove the new offer is universally cheaper: other fees or constraints may exist, and home food may cost less. Keep the occasion and units consistent and record the origin of each number. These figures are invented case inputs, not current market prices.

The matrix reveals three different possible gaps. The cafeteria has a time gap. Existing delivery has an access gap. Home preparation has an effort cost. A venture may address one of these and not the others. Promising “delivery to the ward” before permission is obtained turns an operational assumption into an unsupported commitment. If bringing a meal works well for a worker, do not label it a failure simply because it generates no venture revenue. A successful habit can be the strongest alternative.

To turn the comparison into research, select one possible gap and ask about a recent incident. “On your most recent shift when you wanted food after midnight, what did you do, and how long did it take?” is more useful than “Do you lack healthy options?” Ask about time, location, approximate expense, and what was acceptable or frustrating. The word “better” hides trade-offs: faster pickup may cost more, while a lower price may require home preparation. Ask which two criteria mattered in the actual occasion rather than asking for a general vote on an imagined product. If answers conflict, preserve the split instead of reporting an average that hides different segments.

As evidence arrives, label each matrix cell according to its basis: a directly checked opening hour, receipt or timed process; a worker's account of a specific incident; or an untested team estimate. These sources do not carry the same weight. Three accounts of a ten-minute walk to the gate indicate an access problem for those people, not its prevalence across all workers. If the site rules prohibit the planned internal pickup point, mark the proposed service unavailable in its present form. The team may change the handoff, the first segment, or the test. Competitive analysis is meant to locate feasible value, not to make every current alternative look bad.