Skip to content
Zenor

How we work · I-REALEvolving method

We start from one real business problem.

I-REAL is the working method we've drawn from practice: clarify the purpose before building, and go back when verification fails.

How we work · I-REAL

  1. Intention

    Purpose

    Ask what you actually want to change.

    • Interviews with owners and real users
    • The purpose behind the request
  2. Representation

    Spec

    Turn business intent into a spec engineers and designers read the same way.

    • Requirements and system assessment together
    • One spec, read the same way by everyone
  3. Execution

    Build

    Build the system, connect the data, or make the film.

    • Proof of concept
    • Development, integration or film production
  4. Assessment

    Verify

    Test it where the work happens. If an assumption fails, go back.

    • Was the purpose achieved?
    • What did it really cost?
  5. Learning

    Improve

    Keep what we learned for the next round.

    • Hands-on support for real users
    • Room to grow with the company

When an assumption fails, we go back and fix it, instead of building more features that miss the point.

For example

  1. The client asks: “Let AI reply to emails automatically.”

  2. Look closer, and the real bottleneck is scattered product data and inconsistent pricing references.

  3. So we build knowledge search, quote drafts and human review first.

We're consultants and engineers at the same time,so whenever we interview a client about their needs,we assess the system in the same conversation.

General Manager, Zenor

Principles

  1. AI does the first pass. People decide.

  2. Deliver the agreed scope; measure after go-live.

  3. Where data lives depends on how sensitive it is.

FAQ

Do we have to start with a big project?

No. We usually start from one real business problem and run a proof of concept that fits your budget and needs.

What if we find the direction is wrong halfway?

We go back to an earlier stage and fix it, rather than keep building features that have drifted from the purpose.

Is I-REAL a proven standard?

No. It's how we organize what we've seen in practice, and it is still being developed and tested on real projects.

You don't need to understand complex AI. Bring the problem you want to fix.