Why does entering just a job title produce more recommendations, not fewer?

This is expected behavior, not a bug

A job title without a specific task creates an ambiguous audience model. The pipeline resolves ambiguity by broadening its guidance scope to serve more possible reader types. The result is more recommendations, not fewer. This is expected behavior.

The numbers

Three conditions compared by combined toClarify and toMakeExplicit guidance items:

"Marketer, Intermediate level" (role and level combined, no task): 8.0 items. No survey at all: 6.8 items. Role, level, and a specific task: 4.0 items.

The study condition described above used both a role and a level together, not a role alone. That combination still produced more guidance than leaving the survey blank. Adding audience information without a specific task widens scope rather than narrowing it. Only a complete description with a specific task reliably brings guidance below the blank baseline.

Note: these figures cover toClarify and toMakeExplicit only. The toAdd count, which reflects concepts missing entirely from the page, follows a different pattern and is not included here.

Two paths to fewer, better recommendations

Complete the survey: fill in role, level, and a specific task. See the task description guide for how to write a task description that produces the tightest results.

Skip the survey entirely: leave all three fields blank. This produces more targeted guidance than a role without a task.

Either path produces a better result than the partial state. If you have already entered only a role, re-running Phase 2 with a complete audience description or no audience description will adjust the guidance. Re-running Phase 2 uses one analysis credit.