Alexey Shurov.
Free tool

What to automate first, a scoring tool

Pick a task your team repeats every week and answer five quick questions. You get a plain answer on whether it is worth automating, what helps and what to watch. Score up to four tasks and it tells you where to start, and you can keep the result as a private plan page.

How the score works

Each task gets points for five things. How much of the week it takes, whether it arrives in the same format each time, whether the rules are written down, whether you keep past examples to test against, and how easy a mistake is to catch. The most a task can score is 14.

The questions are the ones that decide whether a project works. Volume makes it worth doing. A regular format and written rules make it possible. Past examples make it testable. The cost of a mistake decides how much a person stays involved.

What the answers mean

A strong place to start is frequent work that follows rules and has past examples to test against. Run it alongside the team first, so its work can be compared with theirs before it acts on anything.

Keep a person deciding means the work is mostly judgement. The AI gathers the facts and suggests, and a person makes the call.

Too small to be worth it means the task takes under two hours a week, so a form, a template or a simple script will fix it for less. Collect examples first means no past examples are kept, and without real ones there is no honest way to test what you build.

The plan page

Leave a work email under your answer and I make you a private page. It has the first steps, what the work costs you today, a short list of what I would need from you, and a note from me once I have read your answers.

There is no password. Your link is the way in. You can send a read only copy to a colleague, and you can delete the page at any time.

Using it with an AI assistant

If you work with an AI assistant, it can ask you the same five questions in a chat. Assistants that connect to tools can use the free one at shurco.ai/mcp, which scores the task the same way and gives back a link.

The link opens your answer on this page. Nothing reaches me until you ask for a plan page yourself.

What it leaves out

It is a first sort, not a design. It does not know your systems, how clean your data is or how the team feels about the work, and each of those can change the order. Use it to choose what to look at properly.

Your answers stay in your browser unless you ask for a plan page. The site only counts, with no name or address, that a set of questions was finished.

Common questions

What should a business automate first with AI?

Work that is frequent, follows rules someone could write down, arrives in a fairly regular form and has past examples to test against. Order entry, timesheet checking, quote preparation and report compilation are common first choices. Avoid starting with rare tasks or work that is mostly judgement.

Which tasks should not be automated with AI?

Tasks that take very little time, tasks where nobody can state the rules, and decisions where a mistake is costly and hard to catch. For the last kind, use an agent to prepare the decision and keep a person making it.

How do I know if a task is big enough to automate?

Measure it for one real week. If the team spends only an hour or two on it, a simpler fix such as a form or a script is usually better value. The ROI calculator on this site works out the payback for larger tasks.

Is the scoring tool free?

Yes. It runs in your browser. If you want to keep the result, leave an email and you get a private plan page, which is also free.

Want the real numbers for your process

I build and run production AI agents that take repetitive work off operational teams. Send me the task and I will tell you honestly whether it is worth automating.

shurco.aiGuidesFree toolsSolutionsInsightsWork with meRSSllms.txt