Name the question before assigning a number.
A score can appear precise while answering the wrong question. A website assessment describes what a researcher can observe online. A client-feedback measure describes a particular feedback sample. A transaction measure describes activity within a defined period. None automatically establishes the best representative for every property.
Before comparing a ranking, ask what was measured, which candidates were considered and which were not. A selective research set can be useful, provided its coverage is visible. Calling it a complete market ranking would be a different claim.
Missing information is not a poor result.
An absent data point should remain unknown. Treating missing information as zero can penalize a practice simply because it publishes less, while filling a gap with an average creates evidence that was never observed.
The same problem appears when team totals and individual results are mixed. A comparison needs a consistent unit, geography, property category and measurement period. The limitations should be visible next to the result, not hidden behind a confident headline.
Separate collection, interpretation and approval.
Our research workflow separates source collection from assessment and practitioner review. Automated checks can organize records and flag changes. An assessment still needs a defined rubric and supporting evidence. A human-review label requires a recorded approval of the version being published.
The implemented REPR scoring gate does not accept a team name, preferred position, payment or publisher affiliation as an input. It returns no public score when an eligibility requirement, dimension-level observation or version-matched human approval is missing.
What this edition can support.
The current team edition provides identity and service-reference profiles, official-source links, local interview questions and a same-market comparison tool. Complete, comparable performance packets and individual human sign-offs have not been recorded for these profiles, so numerical performance scores remain unpublished.
That is a boundary on the evidence, not a statement that the listed teams are good or bad. As research becomes more complete, the relevant change is the addition of traceable information—not simply a more impressive-looking badge.