Home · Business · Product & Strategy · UX & prototyping

YES

As of 13 August 2026, AI can find friction in your user journey.

This still needs a person who signs their name to it.

Can you do it?

15 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ita colleague

What the alternative costsThe supplied tool information gives no comparable human-service price.

If this goes wrong: the team prioritises a plausible-looking problem while missing the real cause, and spends product time fixing the wrong part of the journey.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice

  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Open the current journey map, prototype or product flow and copy the journey stages, user goal and key constraints into a working document.
    2. Gather the relevant evidence for the same journey, including analytics exports, usability notes, support tickets, survey responses and observed session problems, and remove personal data that the analysis does not need.
    3. Paste the journey details and evidence into the prompt, keeping each source labelled so the model can distinguish observation from interpretation.
    4. Ask the model to produce the friction table, alternative explanations and three validation checks, and reject any finding that has no reference to supplied evidence.
    5. Compare each suggested problem with the underlying analytics, user quotes or session observations, then mark whether it is confirmed, contradicted or still a hypothesis.
    6. Ask a product or UX colleague to challenge the three highest-priority findings, then record the validation test and owner before changing the journey or roadmap.

    Prompt

    You are a UX researcher helping me find friction in a user journey. Analyse only the evidence I provide and do not invent user behaviour, causes, frequencies or business impact.

Product or service: [NAME]
User group: [WHO THESE USERS ARE]
Journey goal: [WHAT THE USER IS TRYING TO ACHIEVE]
Journey stages: [PASTE THE STAGES IN ORDER]
Available evidence: [PASTE ANALYTICS, USER RESEARCH NOTES, SUPPORT TICKETS, SESSION OBSERVATIONS, SURVEY RESPONSES OR USABILITY TEST NOTES]
Known constraints: [TECHNICAL, LEGAL, COMMERCIAL, ACCESSIBILITY OR OPERATIONAL CONSTRAINTS]

For each suspected friction point, provide:
1. The exact journey stage.
2. A plain description of what may be difficult or confusing.
3. The evidence supporting it, quoting or referring to the supplied material.
4. What is directly observed versus what is an inference.
5. Possible alternative explanations.
6. The users most likely to be affected.
7. A suggested severity and confidence rating, with a short reason for each.
8. The cheapest practical research or product check that could confirm or reject it.

Separate evidence-backed findings from hypotheses. Do not recommend a solution until the friction has been validated. End with the three checks I should run first, in priority order, and explain what result would change the priority.

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot observe unrecorded behaviour, hesitation or workarounds in the real product.
  • AI cannot tell whether a recurring complaint is a root cause, a symptom or a request from an unrepresentative user without further research.
  • AI cannot reliably judge the commercial, accessibility or organisational importance of a friction point from a journey map alone.
  • AI can over-weight clearly written evidence and under-weight silent users who are missing from the dataset.
  • AI cannot take responsibility for prioritising a fix or for the effect of a wrong intervention on customers.

Even on a YES, the friction has a name: context depth, judgement under ambiguity and verification cost.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs2
Verification1
Liability2
Effort delta2
Total9 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT analyse my user journey?
Yes. Give it the journey stages and labelled evidence such as analytics, research notes, support tickets or usability observations, and ask it to separate findings from hypotheses. It can organise the evidence and suggest validation checks, but it cannot see behaviour that you have not supplied.
What data does AI need to find UX friction?
Provide the user goal, journey stages, relevant user group, analytics or funnel evidence, research observations and known constraints. Remove unnecessary personal data and label each source so the model does not present an inference as an observation.
Can AI tell me what to fix first in my user journey?
It can propose a priority using the evidence and your stated criteria, such as affected users, task importance and confidence. You and a product or UX colleague still need to validate the problem and decide what to fix first.
Is AI user journey analysis accurate?
It is useful for finding patterns and gaps in the material you provide, but its conclusions are only as good as that material. Check each suggested friction point against the underlying evidence and run a focused usability or product check before treating it as a confirmed problem.

Nearby answers

All UX & prototyping rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.