PROOF I count the failures

ioMoVo Mylo AI NXNE Festival Roots Canada McDonald’s app Toronto Cupcakes

Six problems, and what the research changed

If a project has no metric, its proof says so.

All six, most recent first

  1. 01 ioMoVo 2025 · Toronto Metropolitan University

    A platform that could search petabytes with plain English, and nobody could find the switch that turned it on.

    Outcome · 6 of 6 missed the relevance toggle

  2. 02 Mylo AI 2025 · 14 weeks

    Parents at 3am do not need more information. They need the answer without the forum thread.

    Outcome · Answers traceable to the parents who wrote them

  3. 03 NXNE Festival 2024 · 7 weeks

    Hundreds of artists across twenty-plus venues, and no app. Nobody wanted the full schedule.

    Outcome · Set conflicts got the strongest response in testing

  4. 04 Roots Canada December 2023

    Sales down 17% in three months. Traffic flat. The new checkout had tested well. Nothing looked wrong.

    Outcome · A 10% lift per change, named before anything ran

  5. 05 McDonald’s app 2024 · 8 participants

    Seven of eight people could not find the favourites list. All eight said they were satisfied with the app.

    Outcome · 7 of 8 never found favourites

  6. 06 Toronto Cupcakes 2024 · Full site redesign

    A bakery known for being warm and generous, with a crowded site and a checkout long enough to lose people halfway through.

    Outcome · Four changes out of three sessions, one of them a removal

How each one is written

Seven beats, in the same order every time

Next

Ask about any of these, including the parts that did not work

The constraints section on every page is there so you do not have to ask what was missing. If you want the rest of it, the fastest route is an email with the project name in the subject line, at the address in the footer.