Your website loses customers every day. This tells you why.
A lightweight survey widget that lives on your site. Describe what you need — AI builds the draft in seconds. Or start from 12 expert templates crafted from 15 years of real A/B tests. Either way, you'll know exactly what to ask, when to show it, and what to do when answers come back.
Why didn't you sign up?
Conversion & CRO
Pricing page clarity check
Conversion & CRO
Homepage clarity check
Messaging & Positioning
Navigation check
Website & UX
What stopped you from signing up?
What stopped you from signing up today?
What stopped you from signing up today?
AI Insight
34% cite pricing. Consider adding a comparison table or rewriting plan descriptions using visitor language.
You have the data. You don't have the story.
Analytics tells you what happened. It never tells you why. That gap is where revenue dies.
The dashboard stare
Traffic is up. Conversion is down. The pricing page has a 62% bounce rate. Your boss asks: why?
You don't know. Because analytics tracks clicks, scrolls, exits - never what visitors think.
The tool graveyard
Hotjar - $99/mo for the one survey feature that matters.
Typeform - sends visitors away from the page you're optimizing.
SurveyMonkey - you opened it, saw the interface, closed the tab.
Google Forms - a tax form on your $50,000 website. Removed within the hour.
The A/B testing wall
Most SaaS sites don't have enough traffic for A/B tests to reach statistical significance. You need months, thousands of conversions, and a result that's still just a p-value.
Surveys don't need stat sig. 50 visitors telling you “I don't understand the pricing” is something you can act on by Friday.
The real problem
Even with the right tool - what would you ask? “How was your experience?” gets nothing actionable. “Any feedback?” gets “looks great!” from the 2% who bother.
Most teams don't know what questions produce insights you can act on. What to ask. On which page. At what moment.
That's the problem we built this for.
The $20K research project
You hire a UX agency. They recruit participants, run sessions over Zoom, spend four weeks synthesising. The report arrives 6 weeks later.
The headline finding: “users find the pricing confusing.” You already suspected that. You just needed data to act on it.
A 3-question survey on your pricing page, triggered on exit intent, answers the same question in a week — from real visitors, in the real moment they decided not to buy.
One script tag. Expert questions. Real answers by Friday.
Selge is a micro-survey widget that lives directly on your website. But the widget isn't the product. The templates are.
Pick an expert template
Don't start from a blank survey. Start from 15 years of web optimization. Each template is a complete playbook: the exact questions to ask, which page to put them on, what trigger to use, how to interpret the results, and what to change based on what you learn.
You're not building a survey. You're deploying a research methodology.
Why didn't you sign up?
Conversion · 3 questions
Homepage clarity check
Messaging & Positioning · 2 questions
Pricing page clarity check
Conversion & CRO · 4 questions
What else are you considering?
Conversion & CRO · 2 questions
What stopped you from signing up?
Make it yours in 2 minutes
Swap a word. Change the color to match your brand. Preview exactly how it looks on your site - live, in the builder.
The defaults are already good. Most people change the color and ship it.
Paste one line, go live
One script tag. That's it. Your survey appears on the right page, at the right moment. Exit intent on the pricing page. Scroll trigger on the features page. Time delay on the homepage.
The targeting is already configured by the template. You just activate.
<script async
src="https://widget.selge.app/v1/loader.js"
data-project="your-project-id">
</script>Not just questions. A complete playbook.
Every template comes from real conversion work: 40+ A/B tests, hundreds of pricing page optimizations, and a decade of watching what actually moves business metrics.
“Why didn't you sign up?”
Conversion · 3 questions · Exit intent trigger
1. What stopped you from signing up today?
2. What would have made you sign up?
Open text response
3. How clear was our pricing?
Analytics tells you where they drop off. This survey tells you why - in their own words.
Exit intent on your pricing or signup page. Catches people who considered buying and decided not to.
“Too expensive” >30%? Usually not a pricing problem - it's a value communication problem. “Not sure it fits” >25%? Messaging problem. Rewrite using visitor language.
Group open-text answers into 3-5 themes. Pick the top 2. A/B test a fix for each. Re-run the survey after changes.
Homepage clarity check
No analytics tool can tell you if visitors actually understood your homepage. This two-question survey can.
2 questions
Pricing page clarity check
Is your pricing page clear - or are visitors too polite to tell you it's confusing?
4 questions
Did you find what you were looking for?
The simplest survey. The most powerful insight. Directly measures whether a page fulfills visitor intent.
3 questions
What else are you considering?
When 35% of visitors are cross-shopping one competitor, you build a comparison page. This survey builds your competitive strategy with zero research budget.
2 questions
Navigation check
Navigation problems are invisible in analytics. Bounce rates don't tell you why someone left. This survey catches friction directly.
3 questions
Was this article helpful?
Time-on-page is misleading. Someone confused stays longer. This gives you a direct quality signal - per article.
2 questions
Your survey data, inside your AI workflow
Query your survey results in plain English and get instant analysis. Works with Claude Code, Cursor, and any MCP-compatible AI assistant.
Claude Code
Cursor
WindsurfReal Claude Code conversation · Selge MCP server
From dashboard confusion to data-backed changes
See the difference when you stop guessing and start asking.
You redesign by gut feeling
Pricing page gets redone because it “feels right.” The CEO saw a competitor do something. Conversion moves 0.3%. Was it the change, the season, or noise?
247 visitors tell you exactly why
Deploy “Why didn't you sign up?” on your pricing page. Within a week, real answers roll in.
Top responses
Changes are shots in the dark
You rewrite copy because someone on the team didn't like it. You add features because a blog post said to. No data. No direction. Just opinions.
You act on real words from real visitors
Add a comparison table. Rewrite plan descriptions using the exact words visitors used. Changes backed by evidence, not opinions.
AI insight
“34% of respondents can't distinguish between Growth and Pro plans. Recommend adding a feature comparison table and clarifying the value difference.”
Three months later, the cycle repeats
Someone asks “should we redo the pricing page?” again. Nobody remembers what changed or why. You start from zero.
The objection drops from 34% to 11%
Run the survey again. Measure the change. Conversion goes up. You know exactly what worked and why.
Survey comparison
Before changes
After changes
You've probably considered these already
vs. Hotjar
You wanted surveys. They sold you heatmaps, session recordings, and a feedback widget. You're paying $99/month for the one feature that matters.
We do surveys. Only surveys. Starting at $19/month.vs. Typeform
Beautiful forms - on typeform.com. To survey pricing page visitors, you'd send them away from your pricing page. That defeats the purpose.
The survey lives ON your website. Visitors never leave the page.vs. Survicate
Good product. Starts at $99/month. The design is functional but not beautiful. The templates don't tell you what to do with the results.
Better design. Smarter templates. A third of the price.vs. SurveyMonkey
Built for 100-question academic research surveys. Not for 3-question on-site micro-surveys on a SaaS pricing page.
Modern. Lightweight. Built for exactly this use case.vs. Google Forms
It's free. It also looks like a government form embedded on your $50,000 website. No targeting. No analysis. No strategy.
Beautiful widget. Smart triggers. AI analysis. Expert templates.vs. DIY / Custom code
"Let's just build a quick survey ourselves." Famous last words. It never gets prioritized. When it does, it breaks on mobile.
One script tag. Maintained for you. Expert templates you'd never think to create.Built by someone who's done this 500 times
Margus Veeber
15+
Years in web
500+
A/B tests
12
Expert templates
🇩🇪
Built in Germany
GDPR
Compliant
I'm currently Head of Web at Pipedrive - a B2B SaaS that scaled from early-stage to a global product used by 100,000+ businesses. I've run 500+ A/B tests, optimized hundreds of pages, and led web strategy through multiple growth phases. Through VeeberMedia - my web advisory practice - I've applied the same approach with SaaS clients and built products, Selge included.
What I learned over and over again: continuous feedback from customers changes everything. Putting a 3-question survey on the right page would surface real objections within days. That tight loop - ask, learn, adjust, repeat - made every other decision faster and better grounded. Not a replacement for analytics. The missing piece that made analytics actually useful.
Every template in Selge comes from that experience. Not theory. Not best practices copied from a blog post. Real patterns from real SaaS websites with real traffic - built to give you that same continuous feedback cycle on your pages.
The survey features that matter. Without the $99/month bloat tax.
The DIY Approach
What you're doing now
Embed a random Google Form on your $50,000 website. Wait 3 weeks for 2 responses. Present “insights” that are just vibes.
The DIY Approach
plus your sanity
What you're doing now
Free to sign up. Free to build. Free to preview.
You only pay when you're ready to go live.
Next week, you could know why visitors leave your pricing page.
Or you could keep staring at the bounce rate.
Quick question
What stopped you from signing up?
Why didn't you sign up?
/pricing · Exit intent · Last 7 days
0
73%
+18%
New response - /pricing
"Can't tell plans apart"
New response - /pricing
"Too expensive"
AI INSIGHT
“34% can't distinguish between plans. Add a comparison table and clarify tier differences.”