Research · Benchmark
PTAS AI 2026 Automation Benchmark
How automation-ready is your business? A short assessment, an Automation Readiness Score out of 100, and what to do about it.
PTAS AI 2026 Automation Benchmark
How automation-ready is your business?
Fourteen questions, about seven minutes, and an Automation Readiness Score out of 100 with the two places we would start.
The benchmark asks how work moves through your operation today: how routine tasks get done, how documents enter your systems, how clean the master data is, how systems exchange data, how approvals run, where AI is already used and who can change a rule. Each answer is scored, the scores are weighted, and the total is your Automation Readiness Score.
We call it a benchmark because it holds every answer against the same fixed rubric, in the same order, every time. It is not a league table. There are no other companies behind your number, and this page publishes no results or statistics from the people who have taken it.
What it is good for is the first meeting. A score of 47 with Document Automation at the bottom is a more useful agenda than "where should we use AI?", and it is an agenda your finance, operations and IT leads can each check against what they see. You take it through the finance workflow assessment if payables is already the candidate, or book a call and work through it with the team.
Seven dimensions
Where the 100 points go
Two questions per dimension. The weights say where readiness is usually won or lost: in the process and its documents far more than in how much AI a team has used.
Automation Readiness Score · how the 100 points are split
Process and documents carry 45 of the 100 points. Governance carries five, and it can still stop a live project faster than any other dimension.
25 points
Process Automation
How much routine work still runs on manual steps, and whether a person, a script or a system picks up a task when it comes round.
20 points
Document Automation
How invoices, receipts and forms get into your systems, and how much of what is read is checked against a PO, master data or tax records.
15 points
Data Quality
Whether vendor, customer, item and account masters are clean and owned, and whether a reported number can be traced to its document.
15 points
System Integration
How the ERP, CRM, banking and document stores exchange data, and what it would take for a new tool to post into the ERP next month.
10 points
Workflow Maturity
Whether approval thresholds are written rules or habits, and whether you find out a case is stuck from an alert or a complaint.
10 points
AI Adoption
Where AI runs today, if anywhere, and what the team learned from automation projects it has already tried.
5 points
Governance
Who can change a rule that affects payments or compliance, how the change is recorded, and how access and retention are enforced.
Two tracks
Same weights, different questions
The finance track keeps the seven dimensions and asks about payables instead: invoice intake, matching, vendor data, posting, GST validation and payment limits.
The arithmetic
How fourteen answers become one number
Four steps, the same for every respondent and both tracks. Nothing is adjusted by industry or company size.
-
01
Answer
Each of the fourteen questions has four options, from the most manual to the most systemised. Pick the one closest to how work runs today.
Typed in by hand → 0 · Extracted, low-confidence fields reviewed → 3
-
02
Average
The two answers in a dimension are averaged and divided by three, giving a share between nothing and all of that dimension.
2 and 1 → 1.5 of 3 → half
-
03
Weight
The share is multiplied by the dimension weight and kept to one decimal place.
Half of Process Automation → 12.5 of 25
-
04
Add
The seven results are added and rounded to a whole number. That number sets the band.
46.7 → 47 → Developing
One result, worked through · Illustrative example
Illustrative data. A team with no AI in production can lose ten points at most. A team that keys documents by hand and checks none of them loses all twenty for documents.
Result bands
Emerging, Developing, Automation Ready, Advanced
The band is a reading of the whole. The seven bars on the result page are where the useful detail is.
Under 40
Emerging
Most work is still manual and disconnected. The first project should be small, document-led and measurable.
40 to 59
Developing
Parts of the operation are ready and parts will fight you. Start where the data and the systems already cooperate.
60 to 79
Automation Ready
The foundations are in place. A focused workflow can go live without a clean-up project first.
80 to 100
Advanced
Automation runs inside governed workflows. The next gains come from widening coverage and tightening the feedback loop.
What comes back
Your two lowest dimensions pick the next step
The result shows the score, the band and a bar for each dimension, and the same result arrives by email with your answers attached. It then points to two pages, one for each of your two lowest dimensions.
Limits
What the benchmark is not
A score of 72 with a vendor master full of duplicates is still a vendor master full of duplicates. The benchmark tells you where to look.
Not a comparison
It does not rank you against other companies
The score compares your answers with a fixed rubric. There are no peer averages behind it and no industry percentile in the result.
Not validated
It has not been tested against outcomes
Nobody has checked statistically whether a higher score predicts a better project. The weights are a considered opinion, written down so it can be argued with.
Not a measurement
It does not know your numbers
Volumes, cycle times, error rates and the cost of the process come from your own queue, measured with your team. The score cannot estimate them and does not try.
The fourteen answers also cannot see your documents. Whether a creased receipt or a forty-line invoice can be read is a separate question, and the DocPro document test answers it on one of your own files. The readiness checklist goes through each dimension in more depth, with what to do about a low one before a project starts.
FAQ
Questions about the benchmark
Is the Automation Benchmark an industry standard?
No. It is PTAS AI methodology. The seven dimensions, the weights and the band thresholds are ours, set from what has decided the outcome of projects we have scoped. It has not been scientifically validated, and it does not compare you with other companies.
What is the difference between the benchmark and the assessment?
The benchmark is the method: what is measured, how it is weighted and what the bands mean. The assessment is where you take it. The general track covers the whole operation; the finance workflow assessment asks the same number of questions about accounts payable. Both use the same scorer.
How long does it take, and what do I need?
About seven minutes for fourteen questions. You need a rough idea of how work runs today, not figures. Name, work email and company are required so the result can be sent to you; everything else on the form is optional.
Do you publish benchmark results or industry averages?
No. This page publishes no aggregate results or statistics. If you tick the research box, your answers may be included in aggregated PTAS AI benchmark research, and any research we publish will state its sample and method. Leave the box clear and your answers are used for your result alone.
My score is low. Does that mean we should not automate?
No. A low score means the first project should be smaller and more document-led, and that some groundwork sits alongside it. The two lowest dimensions tell you what that groundwork is. A low score in a dimension is the first piece of work, not a reason to wait.
Why does governance carry only five points?
Because the weights reflect where readiness is usually decided before a project starts, and that is mostly in the process and its documents. Governance matters more once a project is live, where it is the dimension most likely to stop one. If it is among your two lowest, the result points you to human-in-the-loop design.
Keep reading
Before or after the benchmark
Book a Call
Talk through the seven dimensions with the team, applied to your own processes.
Read more →Assess Your Finance Workflow
The finance track: the same scorer pointed at intake, matching, approvals, GST and posting.
Read more →Automation readiness checklist
The questions behind each dimension, and what to do about a low one.
Read more →Research and Evaluation
The benchmark, the assessments and the document test, and which to use when.
Read more →PTAS AI 2026 Automation Benchmark
Take the benchmark in seven minutes
Fourteen questions, an Automation Readiness Score out of 100, your band, and the two pages we would read first. The copy in your inbox has your answers attached.