Back

Research operations · status discipline

Pilot learning and validation are different achievements

The research path can move from idea to blueprint, methodology, instrument, pilot, data, analysis, reviewed result and validation work. DEPTH-10 makes each transition explicit so a useful pilot is not forced to pretend it proved broad effectiveness.

Prototype and pilot answer different questions

A prototype asks whether something can be built or understood. A pilot asks how a bounded version behaves in a defined context. Neither automatically establishes reliability, validity or broad effectiveness.

Feedback is data, not always outcome

Preference, usability, comprehension, satisfaction, complaint, correction, completion and measured outcome are different classes. Appreciation should not be converted into impact evidence.

Validation needs a defined claim

Ask what construct or outcome is claimed, for whom, under what conditions, with what reliability/validity evidence, comparator, replication or independent review. 'Validated' without scope is too broad.

Negative results still count

A pilot showing confusion, weak completion, no measurable change or a boundary problem still creates evidence. Research integrity does not require every experiment to support the preferred idea.

Reflection

Questions to sit with

  1. What did this pilot actually test?
  2. What result class exists: feedback, completion, measured outcome or validation evidence?
  3. What population/context limits the claim?

One safe next action

Use the Pilot vs Validation Check before publishing effectiveness or validation wording.