Free tool · Automation readiness

Is your business actually ready to automate?

Most automation projects fail before a line of code is written — not because the technology was wrong, but because the process was undocumented, the data was scattered across three spreadsheets and a WhatsApp group, or nobody owned the thing after go-live. This scorecard tests all four of those before you spend a cent.

  • 12 questions
  • About 3 minutes
  • No email needed to see your score
  • Nothing leaves your browser

0 of 12 answered

Part 1 of 4

The opportunity

How much repeatable, rules-based work is actually there. Automation needs volume to pay for itself — a task done twice a month rarely does.

1. How much of your team’s week goes into repetitive admin — re-typing, copying between systems, chasing the same things?
2. The process you would most like to fix — how often does it run?
3. How often does a customer wait on a person for something completely standard — a quote, an invoice, a statement, a booking confirmation, a status update?

Part 2 of 4

Data and systems

Whether the thing can be built at all. Automation reads and writes data — if the data lives in someone’s head or in a chat thread, there is nothing to read.

4. Where does customer and job information actually live — not where it is supposed to live?
5. Could you write the rules for that process on a single page — every “if this, then that”?
6. Can your core systems export data, connect to other tools, or offer an API?

Part 3 of 4

People and adoption

Whether it will still be running in six months. This is where most automation quietly dies, and it has nothing to do with the software.

7. Once the automated process is live, who owns it?
8. How did the last new tool or system you introduced go?
9. If the person who knows this process best were away for two weeks, what happens?

Part 4 of 4

The commercial case

Whether it is worth doing. An automation that saves two hours a month is a hobby, not an investment.

10. Can you put a rand figure on what this problem costs you — overtime, rework, late invoices, deals lost to slow quotes?
11. Is the volume of this work growing?
12. Who signs off the spend, and where are they?

Answer all 12 questions to unlock your score.

The model · Why these four

Four things decide whether automation works

This is not a personality quiz. Each dimension maps to a failure mode we see repeatedly in South African businesses, and each one kills a project in a different way — usually months after the invoice was paid.

1. The opportunity

Automation has a fixed cost to build and a near-zero cost to run. That maths only works when the process runs often. A quoting workflow that fires forty times a week pays for itself; the same build for a process that runs twice a month never will.

The trap is emotional. Businesses want to automate the task that irritates them most, which is usually a rare, messy, high-friction one. Irritation and cost are different things. Score the volume, not the annoyance.

2. Data and systems

Software can only act on information it can read. If your job history lives in a WhatsApp group, your pricing lives in three versions of a spreadsheet, and your customer list lives in someone’s phone, there is nothing to automate against — only something to first consolidate.

This is the most common blocker we find, and the cheapest to fix. Getting one process onto one page and one dataset into one place is usually a fortnight of unglamorous work, not a project. Skipping it is how businesses end up paying to automate a mess.

3. People and adoption

The build is rarely the risk. The risk is that three months later somebody hits an edge case, quietly goes back to doing it by hand, and never tells anyone. Within a quarter you are paying for a system nobody uses.

The single best predictor is whether one named person owns the process after go-live and has the time to own it. “Everyone owns it” means nobody does. The second-best predictor is how the last tool you introduced went — that pattern almost always repeats.

4. The commercial case

You do not need a precise number, but you need a defensible one. If you cannot say roughly what the problem costs in overtime, rework, late invoices or deals lost to slow quotes, you cannot tell whether a solution is cheap or expensive — and you certainly cannot tell whether it worked.

Growth matters as much as size. Automating a shrinking process is a way to spend money on something that is going away on its own. Automating one that is growing faster than you can hire is where the return actually lives.

Reading your score

What each band means in practice

  1. 0–39 · Groundwork first

    Automating now would cement a process you do not fully understand. Nothing here is wasted — the fix is genuinely cheap. Write the process down, get the data into one place, and give it an owner. Then re-take this in a month and watch the score move twenty points without any software at all.

  2. 40–59 · One clear win

    You are ready for one narrow, well-chosen automation. Not a platform, not a transformation programme — one process, automated end to end, proven in production, with a number attached to it. That first win buys the internal credibility you will need for the second.

  3. 60–79 · Ready to build

    You have a defensible case and the foundations to build on. Your real risk now is scope. Businesses at this level typically try to automate four things at once and finish none of them. Sequence the work, ship the highest-frequency process first, and measure it before starting the next.

  4. 80–100 · Ready to scale

    You are past the question of whether to automate. The question is architecture: how the pieces connect, where the single source of truth sits, and how you make the fifth automation cheaper to build than the first rather than harder. This is where a deliberate integration layer starts to earn its keep.

Questions

About this scorecard

Do I have to give you my email to see my score?

No. The score, the four dimension breakdowns and the recommended next step all appear on this page the moment you answer the twelfth question. The form afterwards is entirely optional — it exists only if you want a written reply from us on your specific result.

Where do my answers go?

Nowhere. The scoring runs in your browser with no analytics, no tracking and no storage — close the tab and the answers are gone. The only exception is if you deliberately submit the read-back form at the end, which emails your result to us so we can reply to it.

How accurate is a twelve-question score?

It is a triage tool, not an audit. It is good at telling you which of the four dimensions is your weakest link and whether you should be building or preparing — which is the decision most businesses get wrong. It cannot tell you which specific vendor, integration or model to use. That needs someone to look at your actual systems.

My score is low. Is that a reason not to call you?

It is often the better reason to call. Low scores usually point at process and data groundwork, which is faster and cheaper than a build and makes everything afterwards work. We would rather tell you honestly that you are not ready than sell you an automation that gets abandoned in March.

Is this specific to South African businesses?

The logic applies anywhere; the examples are local. The questions are drawn from what we see most often in South African SMEs — WhatsApp acting as an unofficial system of record, quoting and invoice chasing done entirely by hand, and one long-serving person holding the whole process in their head.

Next step · No obligation

Bring your score to a free 30-minute session

Whatever the number came out at, the useful conversation is the one about your weakest dimension. Bring your result and the one process you keep thinking about, and we will tell you plainly whether it is worth automating yet — including when the honest answer is not yet.