What is AI Readiness Assessment?
Also called AI readiness review.
An AI readiness assessment is a structured review of whether an organization has the data, documented processes, systems access, and governance needed to deploy AI agents successfully. It produces a picture of current gaps rather than a recommendation to buy anything. Findings usually point to unglamorous prerequisites such as undocumented procedures or scattered records.
Assessments generally look at four areas. Data asks whether the information an agent would need is recorded somewhere retrievable rather than held in individual memory. Process asks whether the work has a describable sequence. Systems asks whether the relevant tools have interfaces an agent can use. Governance asks who would be accountable for the output.
The most common finding across published frameworks is that process documentation, not model capability, is the constraint. An agent can only follow a procedure that has been articulated. Work that exists only as a habit in one person's head has to be written down before anything can be delegated to software, which is why assessments frequently trigger a documentation effort first.
Scoring approaches vary widely and none is authoritative. Some frameworks use maturity levels, some use a simple ready or not ready per process, and some produce a weighted score. The number matters less than the inventory of specific gaps underneath it, because that inventory is what actually gets worked through.
Small organizations often skip formal assessment and reach the same conclusions by attempting one process and discovering what was missing. That is a legitimate approach when the cost of a failed attempt is low. Assessment earns its keep when a failed attempt would be expensive, visible to customers, or difficult to reverse.
Key points
- Reviews data, process, systems access, and governance readiness
- Most common gap is undocumented process, not technology
- Scoring frameworks vary and none is authoritative
- Output is a gap list, not a purchase recommendation
- Small teams often substitute a cheap first attempt
In practice
A logistics firm assesses readiness for automated shipment status replies. Data passes, since tracking sits in one system with an API. Process fails, because three coordinators each answer differently and none of it is written. Governance is unclear. The assessment output is not a tool choice but a two week task to write a single agreed reply procedure, after which the process is reassessed.