AI readiness check · free · about 5 minutes
AI readiness assessment and savings calculator
Assess your AI readiness, calculate potential automation savings and discover your first Annex agent. Answer five short sets of questions about your business, systems and one workflow; the report shows your score, the time an agent could free, and where to start.
Start your assessment
Five short steps: your business, systems and data, operations, time and cost, and how you would get started. Most people finish in about five minutes. You can go back at any point without losing an answer.
- A free account is required to see and save the full report. You will be asked to sign in with Google or an email code when you calculate. No password, and no mailing list.
- Until then, your answers stay in this browser only.
- The results are calculated by fixed, published rules. No AI model reads your answers.
Three tools, one assessment
What does the assessment include?
The same answers feed three results. Each one is calculated by published rules, so you can see exactly how it was reached.
AI-native readiness assessment
A score from 0 to 100 across five dimensions: systems and data, knowledge and process, decisions and governance, AI adoption, and how clear the first opportunity is. You get a plain-language classification, your strongest areas and the gaps to close first.
AI automation savings calculator
From the people, hours and hourly cost you enter for one workflow, it estimates the hours an agent could free each year and what that time is worth, in conservative, baseline and optimistic scenarios. Every assumption is shown next to the result.
Find your first Annex agent
It maps the workflow you choose to the Annex agent built for it, says why it fits your answers, and says plainly when that agent is still coming soon, with the agent you could start with today.
Methodology
How the calculation works
Deterministic rules, no AI model: every answer is worth a fixed number of points or feeds a published formula. The tables below are the configuration the calculator runs on.
1. The readiness score
Each answer is worth 0 to 100 points. A dimension’s score is the average of its answers, and the readiness score is the weighted sum of the five dimensions, rounded to a whole number.
| Dimension | Weight | What it measures |
|---|---|---|
| Systems and data | 30% | Whether the records an agent needs exist in systems it can read, rather than in inboxes and spreadsheets. |
| Knowledge and process | 25% | Whether the rules and exceptions of the work are written down, so they can be made machine-readable. |
| Decisions and governance | 20% | Whether limits and approvers are clear enough to tell an agent what it may do and when a person decides. |
| AI adoption and momentum | 15% | How familiar the organisation already is with AI, and how soon it wants to act. |
| Clarity of the opportunity | 10% | Whether there is a clearly defined, largely manual workflow to start with. |
| Classification | Score | What it means |
|---|---|---|
| Building foundations | 0–34 | Much of the knowledge an agent would need still lives in people’s heads, inboxes and spreadsheets. A first agent is possible, but the first weeks go into making one workflow’s records and rules readable. |
| Emerging | 35–54 | Some of the foundations are in place. Pick one well-understood workflow, close the main gaps listed below for that workflow only, and a first agent can follow. |
| Ready for a first agent | 55–74 | Your systems, rules and approvals are clear enough to put one agent to work on one workflow, with people deciding the exceptions. |
| Ready to scale | 75–100 | You have the systems, documentation and governance to run a first agent and add the next ones on the same shared layer. |
2. The savings estimate
hours today = people × hours per period × periods per yearhours freed = hours today × routine share × workload factor × first-year realizationcapacity value = hours freed × fully loaded cost per hour
Periods per year: 220 working days, 46 working weeks or 11 working months. Full-time equivalent: 1,650 hours a year. The routine share is capped at 80% and realization at 90%.
| Workflow | Routine share |
|---|---|
| Answering quote requests and booking orders | 45% |
| RFQs, supplier quotes and purchase approvals | 40% |
| Processing supplier invoices | 55% |
| Following up invoices and payments | 45% |
| Scheduling work orders and deliveries | 30% |
| Handling customer issues | 35% |
| Planning shifts and capacity | 30% |
| Preparing reports and decision briefs | 35% |
| Checking new suppliers or customers | 40% |
| Something else | 25% |
| How manual it is today | Factor |
|---|---|
| A little: most steps are automated | × 0.5 |
| About half | × 0.85 |
| Most of it | × 1 |
| Nearly all of it | × 1.1 |
| Readiness | Realized |
|---|---|
| Building foundations | 45% |
| Emerging | 60% |
| Ready for a first agent | 70% |
| Ready to scale | 80% |
| Scenario | Routine share × | Realization × | Meaning |
|---|---|---|---|
| Conservative | × 0.6 | × 0.85 | Fewer steps handed to the agent and a slower start. |
| Baseline | × 1 | × 1 | The assumptions above, as they stand. |
| Optimistic | × 1.25 | × 1.1 | More of the workflow handed over, adopted quickly. Treat as an upper bound. |
3. Your first Annex agent
The workflow you choose maps to the agent built for it. With “Something else”, the first department you selected that an agent covers decides. Coming-soon agents are marked as such, with the available agent they build on.
| Workflow | Agent |
|---|---|
| Answering quote requests and booking orders | Annex Sell |
| RFQs, supplier quotes and purchase approvals | Annex Buy |
| Processing supplier invoices | Annex Pay |
| Following up invoices and payments | Annex Collect |
| Scheduling work orders and deliveries | Annex Deliver (coming soon) |
| Handling customer issues | Annex Support (coming soon) |
| Planning shifts and capacity | Annex People (coming soon) |
| Preparing reports and decision briefs | Annex Brief (coming soon) |
| Checking new suppliers or customers | Annex Verify (coming soon) |
Limitations
What the results can and cannot tell you
A useful first view, not a verdict. Read the results with these limits in mind.
- It is a structured self-assessment. It reflects your answers, not an inspection of your systems or data.
- The savings figures are estimates from your own inputs and published assumptions. They are not a forecast, a quote or a promise of results.
- Capacity value is time given back to your team, valued at what it costs you. It turns into cash only when that time replaces spend such as overtime, temporary staff, contractors or planned hires.
- It looks at one workflow at a time. A company-wide picture needs one assessment per workflow, or a conversation.
- The agent suggestion follows the scope documented on each agent’s page. Whether an agent fits your systems is confirmed when a deployment is scoped, before it acts.
Questions
Questions about the AI readiness check
Is the AI readiness check free?
Yes. Answering the questions and seeing your full report are free. A free account is needed to see and save the report, so you can come back to it and revise your inputs.
Why do I need an account?
The report is saved to your account so you can return to it, compare revisions and share it inside your company. You sign in with Google or with a 6-digit code sent to your email. There is no password, and creating an account does not subscribe you to any mailing list.
Does an AI model read my answers?
No. The score, the savings estimate and the agent suggestion are calculated by fixed rules that are published on this page. The same answers always give the same result, and no answers are sent to any AI provider.
What happens to my answers?
Until you sign in they stay in your browser only. Once you calculate, your answers and results are saved to your account, and we use them to understand your needs and, if you ask, to discuss them with you. Accounts are deleted 24 months after the last sign-in, and you can delete yours at any time from the results page. The Privacy Notice has the details.
How accurate is the savings estimate?
It is as good as the inputs and the assumptions. The assumptions are deliberately conservative and shown in full, and the three scenarios show the range. Treat the result as a starting point for a conversation, then validate it on your own workflow.
Can I change my answers later?
Yes. From your results page, choose "Revise my inputs". The questionnaire opens with your saved answers, and the report is recalculated when you finish.
My workflow is not on the list. What then?
Choose "Something else". The report then suggests an agent from the departments you selected, or, if none fits, explains how to choose a first workflow. You can always get in touch to talk it through.
What does the readiness classification mean?
There are four bands: Building foundations (0–34), Emerging (35–54), Ready for a first agent (55–74), Ready to scale (75–100). Each comes with a short explanation and the next steps that matter most for your answers.
Prefer to talk it through?
Start with one workflow.
Tell us the work, the systems it touches and who approves today. We will help you choose the first agent and scope it.