Rewriting a complaint into something you can check
The first move is not a proposal. It is rewriting the problem until it can be falsified. Four successive rewrites, and what each one actually adds.
Why the first sentence cannot be worked with
The sentence a client opens with is usually both true and unusable. True, because it describes the discomfort accurately. Unusable, because it names no layer and offers no object anyone could check.
Responding to it with a proposal is replacing parts on a fault nobody has located. So the first move is to rewrite: each pass does one thing, replacing a vague element with one that can be checked, and then asks whether the sentence can now be taken away and verified.
Four rewrites
The example below is constructed to demonstrate the action. It is not any client's words and corresponds to no real company's data.
Original: our overseas market is not doing well. No object, no timeframe, no basis for comparison. Nothing to check.
First: our overseas enquiry volume is below expectation. An object appears. Whose expectation, and below by how much, remain uncheckable.
Second: over the past six months, qualified enquiries from the website and the marketplace store fell short of the internal target. Timeframe and source are in. The definition of qualified, and what the target rests on, are still missing.
Third: over the past six months, enquiries from the website and marketplace store that included a fleet size fell short of the target sales set in January against the same period last year. Now it can be checked. But it states only a shortfall in quantity; it has not located a layer. Is nobody seeing us, or are they seeing us and not proceeding?
Fourth: before those enquiries, can a buyer tell from public pages alone that we are a manufacturer and not a trading company? This one can be falsified by a single page review, and it lands directly on the first gate of the trust chain.
What each pass adds
An object: replace the market with a specific channel, page or buyer role.
A timeframe: give an interval, so the data can actually be pulled.
A definition: state what a word means in this sentence, particularly words like qualified, valid and key, which usually mean different things in different departments.
A basis: state what the comparison is against, who set it and on what grounds. A number of not doing well problems dissolve at this step, because the basis turns out to have no grounds.
A layer: convert the shortfall into a question about mechanism, so it lands somewhere specific. This is the last step and the one that separates a diagnosis from a report.
When to stop
The stopping condition is not that the sentence has become specific enough. It is that an affordable checking action now exists which could prove it wrong.
The fourth sentence can stop because it corresponds to one page review that returns a yes or a no within half a day. A sentence can be more precise than that and still be unfinished, if nobody can verify it at a cost worth paying.
Three ways it fails
The sentence gets longer without acquiring a checkable object. Length is not precision.
The middle steps are skipped and the mechanism is asked about directly, so the question detaches from the numbers the client actually cares about, and the client does not recognise it as their problem.
The rewrite is completed and nobody performs the check, and the conversation moves to solutions. Then all four passes were wasted. The value of the rewrite is that it leads to a check, not that it reads more professionally.
Recent judgements
Registered statements used here
- The first move is not a proposal. It is rewriting the problem until it can be falsified. The opening sentence a client brings is usually both true and unusable: it describes the discomfort accurately while naming no layer and no object anyone could check.[public]
- The stopping condition is not that the sentence has become specific enough. It is that an affordable checking action now exists which could prove it wrong.[public]
- In a client's own account, fact, assumption, constraint and wish arrive inside the same sentence. Until they are separated, every judgement built on top is built on sand.[public]
- An outbound diagnosis does not answer which country or which channel. It establishes under what conditions, by which route, a company enters which markets; which constraints will break that route; and whether the next move is to gather evidence, build capability or start testing.[public]