AI ROI Audit for organisations with 10 to 250 staff
Find where AI paysfor itself first.
Most businesses know they should be using AI. Few know which process to start with. The AI ROI Audit from Concept Informatics finds the one or two places where AI returns the most for the least risk, then proves it on your own data before you commit to a build.
The problem
Why most AI efforts stall
They start with the tool
Licences are bought first and a use is looked for afterwards. The result is scattered experiments that nobody can measure or defend at budget time.
They have no baseline
If today's time, cost and error rate were never recorded, any improvement is an opinion. Finance does not fund a second project on opinion.
The pilot never leaves the pilot
A demo works on clean sample files, then meets real exceptions, real data and real habits. Without a plan for that, it quietly gets dropped.
Free, takes five minutes
Your AI Opportunity Report
Tell us where your team's time goes. You get an instant readiness score and a ranked shortlist with a savings range for each area. To receive the full report, send us your details and we email it to you within one working day.
The scorecard needs JavaScript, which this page uses only for the scorecard and the forms. Without it you can still email info@conceptinformatics.com and we will reply within one working day.
The paid audit
The AI ROI Audit: two weeks, one fixed fee
The free report shows where to look. The audit measures what is really there, ranks every candidate, and proves the best one works before you spend on a build.
Map
Interviews with the people who do the work. We time-map five to eight core processes and record today's hours, error rates and cost. This is your baseline.
Score
Fifteen to twenty-five candidate use cases are rated on annual value, effort, risk and data readiness. The ranking is shown with its workings.
Prove
A working proof for the top-ranked use case, built on your own sanitised data. Output is compared with what your staff produce today.
Plan
ROI model with ranges, a 90-day roadmap with an adoption and training plan for the first project, a build-or-buy recommendation and a risk note. Presented to your leadership team.
Illustrative example of how use cases are scored. Your matrix uses your own processes and measured figures.
You leave with a ranked answer, not a list of ideas
Everything lands on one page: what to do first, what to leave alone, and what each choice is worth in pounds a year, with the assumptions visible so your finance lead can challenge them.
What the audit delivers
- Measured baseline reportMeasured hours, volumes and error rates for each process reviewed.
- Ranked opportunity matrixEvery candidate scored on value, effort, risk and data readiness.
- Working proof of valueThe top use case running on your sanitised data, with accuracy measured.
- ROI modelLow, expected and high cases, with payback period and what the saving is made of.
- 90-day roadmapSequenced actions, owners, costs and the success measure for each.
- Risk and data protection noteUK and EU GDPR position, where your data and suppliers sit and how you would exit each, supplier checks, including the questions to put to any AI supplier, and human review points for the first project.
- What we advise you not to doThe processes and tools we recommend you leave alone for now, with reasons, so your first project goes where it can work.
- Decision pageYour options side by side, with a recommendation and what each would cost, so one person can decide in one sitting.
Demonstrating the return
The number is agreed before the work starts
A return you cannot show is a claim. We fix the measure, record the baseline, and report the result in the same units.
- Pick one measure.Hours per task, cost per unit, error rate or turnaround time. One number per use case, written down.
- Record the baseline.Measured from real work over a fixed period, not estimated in a meeting.
- Run the pilot against it.Same measure, same method, with a person reviewing outputs where accuracy matters.
- State what kind of saving it is.Cash, spare capacity or reduced risk. Time saved only counts as money when it is redeployed or avoids a hire, and the report says which.
| Quote preparation | Before | After |
|---|---|---|
| Quotes a month | 140 | 140 |
| Time per quote | 50 min | 20 min |
| Staff hours a month | 117 | 47 |
| Hours saved a year | 840 | |
| Value at £28 an hour | £23,520 | |
| Running cost a year | £1,200 | |
| One-off build | £7,500 | |
| Payback | about 4 months | |
Fees
Start free. Pay for the next step only when the last one has earned it.
Fees are in pounds sterling and exclude VAT.
Concept Informatics Ltd, VAT number GB 746 4847 93.
AI Opportunity Report
- Readiness score
- Ranked shortlist with savings ranges
- Full report by email
AI ROI Audit
- Measured baseline
- 15 to 25 use cases scored
- Working proof on your data
- ROI model and 90-day roadmap
Proof-of-value pilot
- Build of the top-ranked use case
- Success measure agreed in writing
- Audit fee credited in full
Ongoing support
- Monitoring and tuning
- Next use cases from the roadmap
- Staff training and guidance
We refund the audit fee in full if no use case reaches a projected payback of 12 months or less on the expected case. You keep the report. The terms are in the refund policy.
A smaller business, or not ready for a two-week audit? Start with the AI Quick Start.
Fit
Who this is for
A good fit
- Organisations with 10 to 250 staff
- Teams with repeatable, document-heavy or admin-heavy work
- Leaders who want a number they can take to the board
- Records kept in digital systems, even if they are untidy
Not yet a fit
- Work that is mostly paper with no plan to digitise it
- Looking for a developer to build without measuring the result
- No one available to own the outcome inside the business
Questions
Before you ask
Do we need developers or a data team?
No. The audit is run with the people who do the work and one named owner on your side. We handle the technical side, and we write findings in plain language for non-technical leaders.
What access do you need to our data?
Sanitised samples of the documents and records in scope, under a signed confidentiality agreement. We keep to the minimum needed, record how personal data is handled in the GDPR note, and send nothing to an AI tool without your approval. The terms are in the data processing agreement.
How accurate are the free report's savings figures?
They are planning ranges. They apply typical shares of time that can be automated to the hours you enter, and every assumption is shown. The audit replaces them with figures measured in your business.
What if AI is not worth it for us?
You will have that in writing, with the reasons. If no use case reaches a payback of 12 months or less, the audit fee is refunded.
Why should we trust you with our data?
Ask us what you should ask any AI supplier: what did you do before, and have you moved organisations onto new software and supported them afterwards? Geoff Davison has delivered software into organisations for more than 40 years. We take sanitised samples only, and the risk note records where your data and suppliers sit and how you would leave each of them, including us.
Which tools will you recommend?
Whichever fits your process, security needs and budget. The recommendation can be an off-the-shelf product, a configured workflow or a custom build, and it is justified in the roadmap.
How is the free report different from the audit?
The report is an estimate from your answers. The audit is a measurement from your real work, with a working proof and a plan your finance lead can check.
When do I get the full report?
We prepare it by hand and email it within one working day of your request. The scorecard result on this page is instant, but the full report is not.
Next step
Book your audit
Tell us a little about the business and what you would most like to fix. We reply within one working day to arrange a 30-minute scoping call. There is no charge for that call.
The terms for paid work are in the master service agreement.