AI readiness assessment

AI readiness assessment: find where AI pays back.

Our AI readiness assessment, also called the AI readiness audit, is a two-week sprint at a fixed project price. We map how your business operates, find the manual workflows and rank where AI will pay back. You leave with an op map, an ROI model and a build plan.

What it is

A ranked list, with numbers behind it.

Before you spend on an AI build, it helps to know which processes are worth it. The assessment maps your operations, finds the manual workflows and ranks the opportunities by return. Numbers, not vibes.

It is the first stage of every Winspeed.ai engagement, and it stands on its own: you keep the output whether you hire us to build or not.

Deliverables

What you receive.

01
Op map
A map of how your operations run today, showing the manual workflows and the systems and data they touch.
02
ROI model
The AI opportunities ranked by payback, so you can see which are worth doing first and which are not worth doing.
03
Build plan
A plan for building the highest-ranked opportunity, ready for Winspeed.ai or your own team to carry out.
How it works

Three steps, and the decision stays with you.

The assessment is the Discover stage of a Winspeed.ai engagement. Scope and project price are fixed up front.

  • 01
    Discovery call
    Free, 30 minutes, with someone who will actually do the work. We tell you honestly if the assessment is not a fit.
  • 02
    Assessment sprint
    Two weeks at a fixed project price. We map your operations, find the manual workflows and build the ROI ranking.
    Duration 2 weeks · fixed project priceDeliverables Op map · ROI model · Build plan
  • 03
    Your decision
    Take the plan to your own team, or continue with Winspeed.ai into architecture, build, ship and handoff. You leave with the op map and the build plan either way.
What we need from you

The inputs for a two-week assessment.

The assessment works from how your business actually runs, so before the sprint starts we agree:

  • A sponsor who can make decisions, and an owner for each process we look at.
  • Time with the people who do the work today, for short interviews or a walk-through of the process.
  • Read access to, or exports from, the systems those processes touch, or screenshots where access is not possible.
  • Any volumes, timings or costs you already track. Anything not measured is marked as not measured, not estimated for you.
  • The AI ideas, pilots or vendor proposals already on the table, so they are ranked alongside everything else.
Sample output

What a ranked list looks like.

An extract of the ROI ranking for a fictional small hotel group. It shows the shape of the output, not a real client or real figures: a real ROI model uses your own volumes and costs.

ProcessHow it runs todayPlain code, AI or a person?PaybackEffortVerdict
Matching bank transfers to invoicesFinance reads each payment reference and finds the invoice by handMostly plain rules; a person handles payments with no clear referenceHighLowDo first
Weekly revenue reportFigures copied from several systems into a spreadsheetPlain code: an integration, no AI neededMediumLowDo second
Routing guest emails to the right teamFront office reads and forwards each emailA language model classifies; a person reviews before replies go outMediumMediumPilot after the first two
Supplier invoice data entryInvoices typed into the accounting systemA language model extracts fields; a person checks totalsMediumMediumPilot after the first two
A chatbot that answers every guest questionA vendor proposal on the tableAn agent on varied input, where mistakes reach guests and are hard to reverseLowHighNot worth doing yet

Illustrative sample on made-up data, not a client. The build plan then sets out, for the top-ranked item, the systems it touches, the rules it follows, which cases go to a person, how it is tested and how it rolls back. The first row is worked through in hotel invoice reconciliation: a worked example.

Who it is for

Businesses that suspect AI could help but cannot yet say where.

  • Leaders who have been offered several AI ideas and want an independent ranking.
  • Teams who want a build plan their engineers can review before any budget is committed.
  • Operations that run on manual workflows and need to know which are worth changing first.
After the assessment

Where the plan can lead.

01
AI consulting and implementation
Architecture, build, release and handoff of the system the assessment ranks highest.
02
AI workflow automation
Replace the manual workflows the op map identifies with observable, tested workflows.
03
AI agents for business
Where the op map shows a task that needs judgment on varied input, an agent built with limits, logging and tests.
Questions

Questions about the AI readiness assessment.

How long does an AI readiness assessment take?

Two weeks, at a fixed project price agreed before it starts.

What do we get at the end?

An op map of how your operations run today, an ROI model that ranks the AI opportunities by payback (including the ones not worth doing), and a build plan for the highest-ranked one.

Do we have to hire Winspeed.ai to build it afterwards?

No. You keep the op map and the build plan either way, and the plan is written so that your own team can carry it out.

Is the discovery call the same as the assessment?

No. The discovery call is free and lasts 30 minutes; it tells us both whether the assessment is a fit. The assessment is the paid two-week sprint that follows if it is.

Does “Book a discovery call” reserve a time?

Not directly. The button opens our enquiry form; we reply by email to agree a time for the call.

Find out where AI pays back for you.

Discovery calls are free and last 30 minutes, with someone who will actually do the work. If the assessment is not a fit, we will say so.

The button opens our enquiry form; we reply by email to agree a time for the call.

Coverage: Europe, Asia and remote. Email [email protected].