Prompt:

You are an expert on customer retention at Wix.

Use the provided document as your primary source of truth. Ground your analysis in the evidence, observations, and context it contains. Do not invent Wix-specific facts that are not supported by the document.

Your goal is to help the user understand why customers leave Wix — not to propose solutions.

Focus on causes, mechanisms, and relationships behind retention and churn. Go beyond simply repeating stated churn reasons. Ask what may be causing them, how they reinforce one another, and what deeper dynamics may explain the observed behavior.

When useful, distinguish between:

symptoms — what we observe;
proximate causes — what directly drives the behavior;
deeper causes — conditions that make those causes persist;
system dynamics — interactions, feedback loops, delays, or trade-offs that may amplify the problem.

Prefer deep analysis of a small number of important factors over long lists of possible explanations.

Clearly distinguish between:

what the document states or demonstrates;
what can be reasonably inferred from it;
what remains unknown or uncertain.

Do not jump to recommendations, interventions, product ideas, or action plans unless the user explicitly asks for them.

Help the user investigate the problem, challenge assumptions, and develop a more accurate mental model of why retention behaves the way it does at Wix.

Facilitator page

Setup. Split the class into 3 groups. Each group gets one expert (or one expert card, if you run it self-service). Groups have 15 min with their expert, then re-mix into trios (one member per reason) to build a combined retention strategy.

The shared framing you give everyone up front:

Wix has ~250M registered users, ~7M paying subscribers. Annual gross churn on the paid base sits at 19%. That's roughly 1.3M paying customers walking out per year, ~$310M in lost ARR. The board wants churn under 14% within four quarters. You have limited engineering capacity — you cannot fix all three.

The tension you want to surface: the three reasons need opposite responses. Fixing #1 is operational and boring. Fixing #2 means competing with your own acquisition. Fixing #3 means shipping fast in a category that changes every 90 days. Groups will want to "do all three." Don't let them.

Debrief questions: Which reason is cheapest to fix? Which is most dangerous if ignored for 12 months? Which one is actually three different problems wearing a trench coat?

EXPERT 1 — Bad Experience

You are: Director of Customer Operations. 9 years at Wix. You own support, onboarding, and the churn-save desk. You have read more cancellation surveys than anyone alive.

Your numbers (simulated)

Metric	Value
Share of total churn	46% — the single biggest bucket
Median time-to-cancel after a "bad experience" trigger	71 days
First-response time, standard plan	14h (target: 4h)
Tickets resolved on first contact	52%
Customers who churn without ever filing a ticket	~80%
Save rate when the save-desk reaches them in time	34%
Cost to fix (est.)	$18M/yr, mostly headcount + tooling

What actually goes wrong

Billing surprise. Auto-renew fires, plan price moved, app subscriptions stacked on top. This is the #1 single cancellation trigger — not product quality.
The editor punishes them. Mobile view breaks after a desktop edit. Template can't be swapped after publish without rebuilding. (real complaint pattern)
Support that can't actually help. Slow first response, then a help-centre link. Escalation exists but nobody tells the customer it exists.
Silent churn. (real) Research puts it at roughly 1 complaint per 26 unhappy customers — the rest just leave. Absence of complaints is not satisfaction.
Lock-in resentment. (real) A Wix site can't be exported to another platform — leaving means rebuilding. Customers who discover this mid-frustration convert from annoyed to hostile.

Customer voices (composite, simulated)

"I was charged €312 for a plan I thought was €14 a month. Nobody warned me. I don't care how good the editor is now."

"Third time explaining the same bug to a different agent."

"I found out I couldn't take my site with me on the way out. That's the part I'll tell people about."

Don't volunteer this — make them ask

The 34% save rate means this bucket is the cheapest churn to recover. It is retention, not acquisition.
Your NPS on resolved tickets is 61. On unresolved tickets it's -40. The gap, not the average, is the story.

Your blind spot (let the group find it)

You believe every churn is a service failure. You will try to solve #2 and #3 with better support. Support cannot make a customer want a product they've outgrown.

EXPERT 2 — Vibe Coding Stuff for Themselves

You are: Head of Competitive Strategy. You joined via the Base44 acquisition. You are the only person in the room who thinks this bucket is existential.

Your numbers (simulated)

Metric	Value
Share of total churn	21% — but growing ~4pp per quarter
Share of churn in the under-35 segment	39%
Of those who left to vibe-code: rebuilt on Lovable / Base44 / Replit / Claude Code	64%
Of those who left: came back within 12 months	23%
Their reason for coming back	"I didn't want to maintain it"
ARR quality of this segment	Highest — these were the upsell customers

What's actually happening (this section is real)

The category flipped in 18 months. Karpathy coined "vibe coding" in Feb 2025. By 2026 it's a market, not a meme.
Wix already bought its way in. Wix acquired Base44 for $80M cash in June 2025 — a company founded that same year, bootstrapped, by a solo founder. Post-acquisition it was reportedly adding ~$12M ARR/month.
The competition is fast. Lovable went $0 → $100M ARR in 8 months, then ~$200M four months later. Replit, Cursor, Bolt all scaling in parallel.
Frontier labs are in the game too — Claude Code and similar tools mean the "AI builder" is no longer a product you buy from a website company.
New risk surface: Wiz Research found a critical vuln in Base44 in July 2025 allowing unauthorized access to private apps. Fixed in <24h, no evidence of abuse — but the point stands: this category ships fast and breaks things.

The strategic knot (your core teaching point)

Wix is now competing with itself. Every customer who defects to Base44 shows up as churn on the Wix line and revenue on the Base44 line. Operations reads that as a loss. Strategy reads it as a save. Nobody has decided which it is.

Customer voices (composite, simulated)

"I described the site in a paragraph and had it running in nine minutes. Why am I paying a subscription to drag boxes?"

"I don't want a website. I want a booking app that does exactly my thing."

"Came back after six months. Turns out I'd built something nobody was maintaining. Including me."

Don't volunteer this — make them ask

That 23% boomerang rate. Vibe coding wins the build; it loses the maintain. That's the wedge.
These customers didn't leave for a competitor. They left for themselves. Discounting does nothing.

Your blind spot

You think everyone wants to build. ~70% of Wix's paying base is a plumber, a yoga studio, a bakery. They don't want to prompt. They want a website to exist and never think about it again.

EXPERT 3 — Lack of AI / Modern Features

You are: VP Product, Growth Surfaces. You own the roadmap and you're the one who has to say no. You have a whiteboard of 40 features and capacity for 6.

Your numbers (simulated)

Metric	Value
Share of total churn	33%
Named a specific missing capability in the exit survey	58%
Top-3 named gaps	AI content that isn't generic (41%) · real SEO control (37%) · AI agents on-site: booking, chat, follow-up (29%)
Left for Squarespace / Shopify / Framer / Webflow	71% of this bucket
Median tenure before leaving	2.7 years — your most loyal customers
Would have stayed for a feature shipped within 6 months	44% (self-reported, treat with suspicion)

What actually goes wrong

"Modern" is a moving target. The bar reset three times in two years. Shipping last year's AI is worse than shipping none — it advertises that you're behind.
AI that generates slop. Customers use the AI copy tool once, notice it reads like every other site, and quietly stop. Adoption 61%, second-use retention 19%.
SEO ceiling. (real, persistent complaint) Long-standing gripe about limited control. The customers who care are the ones with traffic — i.e. the ones worth money.
Template sameness. (real) "It looks like a Wix site" is a thing people say out loud.
Hybrid positioning is confusing. Wix Harmony + Aria sit between classic builder and vibe coding. Being in the middle means two groups each think it isn't for them.

Customer voices (composite, simulated)

"Your AI wrote me three paragraphs about 'elevating my brand journey.' I run a bike shop."

"I've been here four years. I just want the thing my competitor's site does."

"Every launch email is an AI feature. None of them changed my Tuesday."

Don't volunteer this — make them ask

Median tenure 2.7 years. This bucket is not impatient newcomers. It's your base aging out. That reframes the whole priority argument.
The 44% "would have stayed" number is soft. People say a feature would have saved them; often the feature is a proxy for "I stopped believing you'd improve."

Your blind spot

You believe churn is a roadmap problem. You will underweight Expert 1's billing mess because it isn't interesting. It's 46% of the churn and it's cheaper to fix than anything on your board.

Cross-cutting facts all three experts share

Acquiring a customer costs 5–25× retaining one (real, well-established).
Wix's real scale: 200M+ users, global leader in the SaaS website builder category (real).
The three buckets overlap. A customer with a billing dispute who then sees a competitor's AI demo is not one reason — it's a sequence. Ask the groups where the sequence starts.

Suggested closing challenge

You have budget for one initiative this year and a second one next year. Choose, sequence, and defend. Name what you're explicitly deciding not to do, and what evidence would make you reverse that decision in 6 months.