Workflow diagnostic

Find the one workflow worth automating before you spend on a build.

Praeco maps one repeated workflow, tests the risky part on real files or examples, then gives you a build / no-build recommendation.

If the work is too vague, too risky, or too low-value, we say no before you spend on a tool.

What it answers

Five questions before any build.

Most failed automation work starts with a vague workflow. The diagnostic turns the messy request into a simple operating decision.

01Is this repeated often enough?
02Are the inputs and outputs clear?
03Can a human define “correct”?
04Where should AI help, and where should a human review?
05Is the value high enough to justify a custom tool?

What you get back

A practical yes, no, or prove-first answer.

The output is not a deck about AI. It is a short decision sheet your owner or ops lead can act on.

Sample decision sheet Automation readiness
OpportunityRepeated approval work that stalls in chat and email
Hard partSorting clean cases from items needing review
OwnerFounder / Ops lead
ValueHigh
EffortMedium
RiskLow–medium
Proof testRun samples through the hard step before full build
Review gateHuman approves edge cases
RecommendationBuild only if the proof test separates routine work from owner-review cases reliably.

Good candidates

Workflows worth testing

  • Repeated weekly or monthly
  • Clear past examples
  • Known owner or reviewer
  • Enough volume, delay, or risk to matter

Usually not worth it

Workflows we may reject

  • One-off judgement calls
  • No sample files or outputs
  • No agreed definition of correct
  • High risk with no review gate

Start here

Send the messy workflow.

You do not need a polished brief. Send where the work starts, who touches it, where it breaks, and what a useful output looks like.

What Praeco replies with

1. Not worth automating 2. Prove hard part first 3. Scope a 4–6 week build 4. Keep human review gates
Email the workflow

No mail app? Copy the four prompts and send them to [email protected].