What is Data readiness?
Whether the data can support the decision you want to automate. Usually the binding constraint, and usually not what the client came in asking about.
Readiness is not the same as availability. Data can be complete, accessible and well governed and still be unfit, because a definition changed in a reorganisation and nobody restated the history. No model repairs a figure that was already wrong when it arrived.
The clearest illustration in our own record: on a demand-forecasting engagement, promotions and returns had been corrupting the demand history for years, so every forecast built on it was in effect forecasting the promotion calendar. Repairing the definitions took forecast error from roughly 42% to 19%. The model was the easy half, and that is the argument for a diagnostic over a product.
| On this | Data readiness | Data availability |
|---|---|---|
| The question asked | Can this data support this decision? | Can we get to this data? |
| What a clean warehouse proves | Nothing on its own - definitions still have to be right | That access is solved, which is the easier half |
| Typical finding | A named field, in a named system, with a named owner and a date | A data-quality score |
- When it is the right answer
- Before any forecasting, scoring or optimisation work. It is the critical path far more often than the modelling is.
- The service line that delivers this
- Forecasting & Planning AI
The forecast is a flat assumption that hides a twenty-point swing.
Start here
If a term here is the one your board paper turns on, ask us about it.
We will send the entry, the evidence behind it, and the honest note about where it does not apply.
You get a reply within one working day, from the engineer who would do the work - not a sales sequence.