For owners whose business works, except for one part

What part of your business is stuck?

Start there, before anyone sells you AI. I'm Madina Umbetova: 20+ years running businesses and commercial teams, and I build the systems that fix them myself.

Pick what sounds like you

Where it usually starts

What I would do

Diagnose first

Count the calls you miss in a normal week, then the jobs they were worth. Pikettli is what I built for exactly this.

See Pikettli ↗
The method

You don't need to launch AI. You need one broken part fixed.

"Everyone is doing AI, so I should too." "I hired an agency, and they never understood my business." If that sounds familiar, the method starts somewhere else: with where your business loses money or time today. Every stage ends with a decision: go, adjust or stop.

  1. 01

    Diagnose

    Find where the business loses money or time today, and put a number on it.

  2. 02

    Redesign

    Decide how that part of the business should run, before anything gets built.

  3. 03

    Architect

    Choose the simplest fix. Sometimes that means no AI at all.

  4. 04

    Build

    A working version you can open and try, tested on realistic cases first.

  5. 05

    Deploy

    It goes into the tools your team already uses, asks before anything important, and can be switched off.

  6. 06

    Prove

    Compare against the starting number, then keep it, adjust it or stop it.

How the six stages work

The writing

One method every week.

Written by an operator, for operators. One method you can actually build, drawn from 20+ years running commercial portfolios and the last year spent building AI tools instead of briefing them out.

All writing

The publication

Subscribe once. Every Sunday, one method you can use.

Substack manages the subscription. The welcome email brings the link to the library: every free resource in one place, and each new one lands there first.

See the free tools and kits

About
Madina Umbetova

I work where commercial reality meets digital delivery.

I'm Madina Umbetova. 20+ years running portfolios and turnarounds across crop protection, healthcare, chemicals, and distribution, from Country GM in Kazakhstan to a $0.8bn global portfolio today, based in Basel, Switzerland.

For years, having an idea was expensive. You wrote a brief, waited in a queue, paid an agency or a consultant, and got back something close to what you meant. That gap closed last year. I now build my own working tools inside Claude on a weekend, against a clear method. The Closed Gap is where I share what I build and how, so senior leaders can do the same: in agriculture today, transferable anywhere, because the problem is the same everywhere.