If you run a SaaS and you’re tired of guessing why users drop off, this is for you. Conversion funnels show you where people bail, what’s working, and where you’re burning money. But tracking and optimizing them can be messy—unless you use a tool that doesn’t fight you every step of the way. Enter Unless, a personalization and optimization platform that’s actually built for SaaS teams (instead of just marketers with too much budget).
This walkthrough will show you how to track and optimize your conversion funnels in Unless—without drowning in dashboards or falling for “magic button” promises. If you want practical steps, honest advice, and the stuff nobody tells you, read on.
1. Get Clear About What You’re Tracking (Don’t Skip This)
Before you even open Unless, take five minutes to write down your actual funnel steps. Not the ones you wish users took, but what really happens.
For most SaaS, the classic funnel looks like:
- Landing page visit
- Signup (free trial or demo request)
- Onboarding step(s)
- Key activation (e.g., first project created)
- Upgrade or payment
Pro tip: Trying to track 12 micro-steps is a waste at first. Three to five real milestones is plenty. You can always add more detail later.
What to ignore: Vanity events (like “clicked help link” or “visited pricing page” for the 8th time). Focus on actions that move people closer to paying or getting value.
2. Set Up Your Base Tracking in Unless
Alright, now log into Unless. The platform is pretty straightforward, but here’s how to avoid common mistakes.
Add the Unless Script
Unless needs its JavaScript snippet on your site to do anything useful. Put it in your site’s <head>—ideally via your tag manager, or just hardcode it if you’re not juggling a million scripts.
Got a single-page app? Double-check that the script fires on every virtual pageview. Unless can usually handle this, but test it. You don’t want to miss tracking because of some React router weirdness.
Define Conversion Events
Head to Unless’s “Events” section. Here’s what to do:
- Create events for each funnel step you wrote down earlier.
- Use clear names, like
signup_submitted,onboarding_complete,first_project_created,plan_upgraded. - Tie these events to actual user actions (e.g., form submissions, API calls, or button clicks).
What works: Set up the events in your app code instead of just on button clicks in the DOM. You’ll get cleaner, more reliable data.
What doesn’t: Relying only on URL changes or page loads. Users do weird things, and modern apps don’t always reload.
3. Build Your Funnel in Unless
Unless lets you chain together your events into a “Funnel” report.
- Go to the “Funnels” feature.
- Add steps in the order users should complete them.
- Set the window (e.g., should a user finish the funnel within 7 days, or does it not matter?).
Pro tip: Start simple. If your trial is 14 days, make your funnel window 14–21 days so you see the full cycle. Later, tighten it up for testing.
Watch out for: Overcomplicating with too many branches or alternative paths. Unless is good at tracking linear funnels—if your flow is super non-linear, map the main path first.
4. Analyze Where You’re Losing People
Now the fun (or pain) begins. Look at your funnel data after a week or two.
- See drop-off rates at each step. If 40% of users never make it past onboarding, that’s your first fire to put out.
- Compare cohorts: Are users from certain channels (ads, organic, referrals) falling off faster?
- Slice by basic properties: device, location, plan type.
What works: Focusing on big leaks first. Don’t chase a 2% drop between steps 4 and 5 if 60% never get past step 2.
What doesn’t: Staring at the funnel every day and panicking over small swings. Give it enough data to be meaningful.
5. Set Up Experiments to Fix the Biggest Leaks
Unless is built for A/B testing and personalization, so use it.
Simple Experiments to Try
- Onboarding flow: Test removing or reordering steps.
- Signup forms: Try shorter forms or fewer required fields.
- In-app prompts: Personalize messages for new users stuck on step 1.
You can launch these changes as “Experiments” in Unless, targeted to users at specific funnel stages.
Pro tip: Only test one thing at a time on a given funnel step. Otherwise, you won’t know what worked.
What to ignore: “Best practice” tweaks like changing your button color—unless you have a strong reason to believe it matters.
6. Measure, Iterate, and Don’t Get Cute
After running an experiment, check the funnel for real improvement.
- Did more users reach the next step?
- Was there any impact on paid conversions, not just form completions?
If it worked, roll it out. If not, move on. No need to force a “learning” out of every failed test.
Honest take: Most experiments are duds. That’s normal. You’re looking for the 1–2 big wins a year, not endless “micro-optimizations.”
Pro tip: Track your experiments and outcomes somewhere outside of Unless (a spreadsheet is fine). It’s easy to lose track of what you’ve tried.
7. What to Ignore and What to Watch Out For
There’s a ton of advice out there, but here’s what matters:
- Ignore “industry benchmarks.” Your funnel will look nothing like Dropbox or Slack, and that’s fine.
- Don’t obsess over tiny improvements. Get the big leaks first.
- Don’t buy into “AI-powered optimization” unless you can see actual results. Unless has some smart features, but it’s not a magic fix.
- Watch out for data holes. If events aren’t firing, you’ll chase the wrong problems.
8. When Should You Bring in More Tools?
Unless covers most funnel tracking and basic optimization for SaaS. But if you need:
- Deeper product analytics (e.g., feature usage, retention curves)
- User session replays
- Complex multi-touch attribution
…you might want to pair Unless with tools like Mixpanel, Amplitude, or FullStory. Just keep your tracking clean—nothing wrecks a funnel like conflicting data.
Summary: Keep It Simple, Iterate Fast
Don’t overthink your funnel. Track the steps that matter, use Unless to spot the biggest leaks, and run experiments to fix them. Most improvements will come from obvious issues, not clever hacks. Set up your tracking right, ignore the noise, and make one change at a time. That’s how real SaaS teams actually move the needle.
Go fix that funnel.