Your lead generation form is the front door to your sales funnel—and in B2B, every qualified lead counts. But most forms are built on guesswork, bad habits, or features nobody uses. If you want more (and better) leads, stop guessing and start testing. This guide is for marketers, growth teams, or anyone tired of watching form fills stall out.
We'll walk through how to use Optimizely to run experiments that actually move the needle, plus what’s worth testing, what’s a waste of time, and how to avoid over-complicating things.
Why Bother Testing Your Lead Gen Forms?
You can read all the “best practices” you want, but the truth is, what works for one B2B business might tank conversions for another. Industry, audience, even your offer—all of it matters. A/B testing with Optimizely lets you see what actually improves your results, not just what “should” work.
Common issues you might be facing: - Low conversion rates despite decent traffic - Lots of unqualified leads - High drop-off after the first form field - Sales complaining about “bad” leads
The fix isn’t always a total redesign. Sometimes, it’s a one-field tweak or a better headline. But you won’t know until you test.
Step 1: Decide What to Test (and What to Ignore)
Before you start wiring up experiments, step back and pick your battles. Not every idea is worth testing. Here’s what’s usually worth your time in B2B:
High-Impact Things to Test
- Number of fields: More fields can qualify leads, but too many and people bail. Test short vs. long forms—but don’t expect miracles if your offer stinks.
- Field types: Dropdowns, multi-selects, or just plain text? Sometimes “Company Size” as a dropdown gets better data; sometimes it’s a roadblock.
- Required vs. optional fields: Can you collect less upfront? Try making phone number optional, for example.
- CTA copy: “Get Started” vs. “Book a Demo”—minor tweaks here can have big effects.
- Form placement: Above the fold, as a pop-up, or on a dedicated landing page.
- Progress indicators: For multi-step forms, does showing progress help or hurt?
- Social proof nearby: Client logos, testimonials, or “trusted by” badges by the form.
What’s Usually a Waste of Time
- Button color: Unless your button is white on a white background, you won’t get much mileage here.
- Tiny UI tweaks: Micro-animations, subtle drop shadows, etc. won’t save a bad form.
- Adding chatbots “just because”: Unless chatbots actually help your users, they usually just distract.
Pro tip: Before you test, ask yourself—will this change actually teach me something useful, or am I just fiddling for the sake of it?
Step 2: Set Up Your Baseline
You need to know where you’re starting from. Dig up your current form’s data:
- Conversion rate (visitors to leads)
- Drop-off per field (where do people give up?)
- Quality of leads (ask sales for honest feedback)
If you don’t have this, set up tracking first. Otherwise, you’re flying blind.
Step 3: Build Your First Optimizely Experiment
Assuming you’ve already got Optimizely on your site, here’s a simple way to get started:
1. Pick a Single Change
Don’t test five things at once. Pick one tweak—like removing a required phone field.
2. Set Up the Variation
In Optimizely: - Create a new A/B experiment. - Version A is your current form. - Version B is the same, but with your change (e.g., phone field optional or gone).
3. Define Your Goal
Be specific. Usually, it’s “form submissions” (conversion). But for longer forms, you might also track “fields completed” or “qualified leads” if you have backend integration.
4. Set Audience Targeting
If you only want to test with certain segments (e.g., U.S. traffic, new visitors), set that here.
5. QA Everything
Double-check both versions. Broken forms waste time and traffic. Fill them out yourself, or better yet, have someone unfamiliar with the site try them.
6. Launch and Wait
Let it run until you have enough data—usually a few hundred conversions per variant, minimum. Don’t stop early just because you’re eager.
Step 4: Analyze and Learn—Don’t Just Chase “Winners”
Look at the results with a skeptical eye:
- Did one version actually increase conversions, or is the difference noise?
- Did lead quality change? More leads aren’t always better if your sales team hates them.
- Did you spot any weird side effects (e.g., more spam, more incomplete forms)?
If you’re not sure, run the test longer or check results by audience segment. Sometimes a change helps one region but hurts another.
Honest take: Most tests are inconclusive or only move the needle a little. That’s normal. The value comes from stacking small wins and learning what matters to your audience.
Step 5: Iterate—But Don’t Obsess
Once you have a winner (or at least, a useful lesson), roll it out. Then pick your next test.
But don’t fall into the trap of endless micro-testing. If your form is already performing well, and your leads are solid, sometimes you’re better off focusing on traffic or your offer itself.
A few ideas for second- or third-round experiments:
- Multi-step vs. single-step forms (sometimes breaking the form up helps)
- Personalization (pre-filling fields for returning users)
- Testing different offers (not just the form, but what you’re offering in exchange)
Common Pitfalls and How to Avoid Them
Here’s where most teams trip up:
- Testing too many things at once: You won’t know what caused the change.
- Ignoring lead quality: More form fills don’t help if it’s all junk.
- Running tiny tests: If you don’t have enough volume, you’ll get misleading results.
- Chasing “best practices” instead of real data: What works for a SaaS in fintech might flop in industrial B2B.
- Letting tests run forever: If you haven’t learned anything after a reasonable sample size, move on.
Pro tip: Document every test, even the flops. The graveyard of failed ideas is where you’ll find your best insights later.
Tools, Tips, and Final Thoughts
Tools That Actually Help
- Optimizely: (obviously) for A/B and multivariate testing.
- Google Analytics: For tracking form completion, drop-off, and traffic sources.
- Form analytics tools: Like Hotjar or FullStory. Heatmaps and session recordings can show you where users hesitate.
- CRM integration: Make sure your form feeds into your CRM so you can track downstream quality.
Keep It Real
- Don’t expect every change to be a home run. Most aren’t.
- Don’t trust “rules of thumb” blindly—test them yourself.
- Don’t make your form a Frankenstein monster of every “growth hack” you read online.
Wrapping Up: Simplicity Wins
Optimizing lead gen forms isn’t rocket science, but it does take discipline. Start with your biggest pain points, test one thing at a time, and focus on what actually helps your business—not just what looks good in a report.
Keep it simple, keep iterating, and remember: the best form is the one your best leads actually use.