During a mission
During the engagement
Our consultants call on the agents at every stage of the CRO cycle. You receive the deliverables, without installing anything.
AI AgentUX & data audit
Discover Mei
CRO cycle · 01 · Idéation
Mei cross-references your replays, your analytics and UX best practices, then has the journey tested by our 4 Synthetic Users personas. You receive the friction map, ranked by impact, each with its level of evidence.
- Analyzes replays, heatmaps and GA4 funnels
- Has your pages tested by 4 personas
- Cites the applicable UX best practices
- Delivers the friction map ranked by impact
Mei
Audit UX & data
Driven by a Welyft CRO expert
Real-world use cases
What Mei can do for you
Scenario 1 · Funnel audit
Mei
Live
You : Audit this site's mobile checkout funnel and produce the friction map.
09:41 · Seen
MeiI cross-referenced the funnel replays, the GA4 exit pages and the checkout UX best practices. Here is the friction map, ranked by impact, each with its level of evidence: UX best practice, Welyft audit pattern or A/B tested learning.Scenario 2 · Runs the 4 personas
Mei
Live
You : Compare these two product page mockups with the 4 personas.
11:08 · Seen
MeiThe Demanding One, the Novice, the Rushed One and the Skeptic reviewed both mockups. Version B holds up better: the frictions noted on A are cited in support. Their verbatims are concise, to be confirmed by a test.Scenario 3 · Targeted best practice
Mei
Live
You : Which UX best practices apply to mobile filters? Cite your sources.
14:26 · Seen
MeiHere are the relevant best practices, rephrased with their reference, and what our audits have already found on comparable list pages. I'm also flagging the tests in the database that addressed this lever.
Illustrative examples: each mission starts from your data and your tools.
Deployment
Agents trained like experts, updated like products.
Each agent is a versioned plugin on GitHub. Three ways to make the most of it.
Via the Welyft platform
Remote, at Welyft
The fleet runs on our platform, connected to your tools with your consent. An expert validates every deliverable.
Internalized within your organization
Within your organization
We build your agents using our method, customize them, then a GitHub plugin marketplace delivers their updates.
.webp)
Watch Mei audit a page and rank its frictions by impact.
Mei, Welyft's UX & data audit agent, cross-references your replays, your analytics and UX best practices to map the frictions in your journeys.
Book a demoIts skills
Analyze, test, prioritize
Skills · Analysis
- Live
Cross-referenced replays and analytics
Mei reads your Contentsquare replays and heatmaps, then your funnels, exit pages and GA4 segments, to pinpoint where and why your visitors drop off.
/mei:analysis
- Live
Checklist by template
Homepage, listing, product page, cart, funnel: each template is scrutinized using patterns and insights drawn from Welyft audits.
/mei:analysis
- In progress
Competitor benchmark
Patterns observed among your competitors help position your journeys. This building block is under construction.
/mei:analysis
Skills · Validation
- Live
4 Synthetic Users personas
The Demanding One, the Novice, the Rushed One and the Skeptic go through your page and each report where they get stuck, with verbatims to back it up.
/mei:validation
- Live
Mockups compared
Two competing mockups? Mei has the personas test them and names the stronger one, with frictions cited.
/mei:validation
- Live
UX best practices cited
Mei searches through 527 complete UX best practices and cites the reference for each, rephrased for your deliverable.
/mei:validation
Skills · Deliverable
- Live
Friction map
Each irritant is located on the journey and ranked by impact, to decide what to address first.
/mei:deliverable
- Live
A level of evidence for each finding
UX best practice, audit pattern to verify, or A/B-tested learning: you know what each recommendation is based on.
/mei:deliverable
- Live
Screen deck and reading
One slide per insight, annotated screenshots and personas with photos, ready for your committee.
/mei:deliverable
Human first
Mei prepares, the Welyft expert decides.
- Deliverable
- Carte des frictions
- The expert decides
- quels irritants traiter
Its safeguards
- Three bases, three levels of evidence: UX best practice, Welyft audit pattern, A/B tested learning.
- An expert review without behavioral data produces no figures.
- Persona verbatims are synthetic: they replace neither a user test nor an A/B test.
- Only a test can confirm that a recommendation works.
Under the hood
What Mei relies on
Trained on
Notre base d'environ 270 A/B tests, les audits UX menés par Welyft, 50 audits CRO et 527 best practices UX
Data used
- Welyft A/B learnings base (around 270 tests)
- GA4: funnels, exit pages, landing pages and segments
- Contentsquare replays and heatmaps
- Screenshots, URL or mockups of the journey
- Welyft UX best practices and audit patterns
Specialties
- /mei:welyft-ux-audit
- /mei:welyft-synthetic-users
It's live : /mei:welyft-ux-audit, /mei:welyft-synthetic-users
In progress : Competitor benchmarking does not yet have a dedicated capability.
Our offers
Offers where Mei is involved
FAQ
Frequently asked questions about Mei
Une question sur votre contexte ? Nos experts vous répondent.
Posez votre questionDo Mei's personas replace a user test?
No. Synthetic Users personas speed up friction detection and the choice between two mockups, but their verbatims are synthetic. A real user test or an A/B test is still needed to validate.
What data does Mei work on?
Your replays and heatmaps, your analytics (GA4), screenshots or mockups of your journeys, the Welyft A/B learnings base and our UX best practices. Without behavioral data, Mei produces no figures.
Can we test Mei on our site?
Yes: the online AI UX audit runs your page through Mei. For a full audit of your journeys, a Welyft CRO consultant drives Mei on your data.
Who decides which irritants to address?
The Welyft CRO expert. Mei prepares the friction map and its evidence; the expert chooses which irritants to address and in what order.
Do agents have access to our data?
Only to the sources you open, with a scope and access defined by contract. Agents work on your tools and your exports; no personal data is copied into a deliverable.
In a few seconds
Mei, from data to deliverable
A prioritized audit, ready to present
UX & data audit
Insights ranked by impact
Carte des frictions
Testée par nos 4 personas : 4 frictions, classées par impact.
Four profiles review the page
L'Exigeante
The Novice
Le Pressé
The Skeptic
Each with its own requirements and breaking point
527
UX best practices available for consultation
4 analysis pillars, M-I-F-A-PV framework
Checklists by template and by vertical
RICES score
Every insight quantified to decide
An insight format
From proof to recommendation
Deck and document in Welyft colors
Illustrative demonstration: the examples are not client results.
- A prioritized audit, ready to present : UX & data audit, Insights ranked by impact
- Carte des frictions : Testée par nos 4 personas : 4 frictions, classées par impact.
- Four profiles review the page : Each with its own requirements and breaking point
- 527 UX best practices available for consultation
- 4 analysis pillars, M-I-F-A-PV framework
- Checklists by template and by vertical
- RICES score : Every insight quantified to decide
- An insight format : From proof to recommendation
- Deck and document in Welyft colors