Ask the next missing question
When a request cannot yet be evaluated, ask for the specific missing information instead of returning a generic rejection.
- Evidence
- Established practice. Widely and independently practised across commerce. Nothing here is a Specify discovery; these are the ideas most likely to be uncontroversial and least likely to be exciting.
- Difficulty
- Moderate. Difficulty is an initial editorial assessment of how much work an idea usually takes, not a guaranteed estimate for your systems. A change that is small in one storefront can be significant in another because of where the data lives.
The commercial problem
A customer describes what they need and the system returns nothing: no results, an empty configurator state, or a contact form. The business has not decided it cannot help. It has decided it does not have enough information, and then presented that as a refusal.
Why it matters
A person who is told no leaves. A person who is asked a question answers it. The two responses cost the same to build and differ entirely in what happens next, because one ends the conversation and the other continues it with more information than it started with.
Where it applies
- Configurators that can reach an invalid combination
- Search and filter results that can return nothing
- Enquiry forms that accept free text and reply with an acknowledgement
- Any journey where the answer depends on a dimension, a material or a quantity the customer has not stated
Suggested implementation
- List the facts an answer actually depends on, in the order they change the outcome.
- For each dead end, identify which of those facts is missing rather than which rule failed.
- Replace the empty state with the single most useful next question, not the whole form.
- Accept an honest 'I do not know' and route it to a person rather than blocking.
- Show what has been established so far, so the customer can see the conversation progressing.
Risks and constraints
- Asking questions the business cannot act on is worse than asking none, because it implies a capability that does not exist.
- A long sequence of questions before any indication of feasibility reads as an interrogation.
- Question order matters: asking for contact details first converts the whole thing back into a lead form.
What to measure
- The proportion of journeys ending in a stated outcome rather than an empty state
- How far into the sequence people get before stopping
- Whether the enquiries that reach a human now arrive with more usable information than before
How this relates to Specify
This is the missing-information and next-question logic that the complimentary Enquiry Audit already runs on a single submitted enquiry.
When an engagement is worth it
Worth an engagement when the questions depend on production rules that are not written down anywhere, which is the usual case.
Related ideas
Start with evidence
Or start with one of your own enquiries
The complimentary Enquiry Audit takes one real customer enquiry and shows what a structured reading of it produces. It runs on this site and costs nothing.