What's normal for your kind of page
Start with what you're asking people to do, because a free account and a sales call are very different requests.
| Your page asks for | Normal share of visitors | Strong |
|---|---|---|
| A free account that never expires (freemium) | 5% to 15% | Judge it by how many later pay |
| A free trial, no card | 4% to 8% | 12% or more |
| A free trial, card required | 2% to 2.5% | Judge it by how many pay after the trial. About 50% is typical |
| A demo or sales call | 1% to 4% | About 5% |
These ranges come from a few industry reports, including First Page Sage's numbers on 86 software companies. They don't agree exactly. Use them to spot a real problem, and don't lose sleep over half a point.
Why the numbers you hear don't match
You may have heard 2.35%, 6.6% or 3.8%. Each one is honest enough. They just measure different things.
| The number | What it actually is | Should you use it? |
|---|---|---|
| 2.35% | An average across thousands of Google Ads accounts, in every industry | Only for a very rough feel for paid search |
| 6.6% | The typical page out of 41,000 landing pages from Unbounce, with goals from a purchase to a download | No. Too broad. |
| 3.8% | Unbounce's typical software page, mixing demos, trials, downloads and webinars | As a loose starting point |
If you're going to plan around a number, find out what it counted first.
Work out your own number properly
The math is simple: sign-ups divided by visitors, times 100. If 1,000 people visit your trial page and 60 start a trial, your rate is 6%.
Getting a number you can trust takes a little more care.
Measure one page and one action. A rate for your whole site tells you very little if it mixes your blog, your login page and your pricing page. Pick the page you're sending people to and the action it asks for.
Make sure the action is really being counted. In Google Analytics 4, your sign-up or lead has to be marked as a Key Event before you get a rate for it. Then fill in your own form and check that it shows up once. A broken tracking tag can look exactly like a bad page.
Split it by where visitors came from. Most people skip this step, and it often answers the whole question. Here's what Unbounce found on software landing pages:
| Where visitors came from | Typical share who sign up |
|---|---|
| 16.9% | |
| Google search ads | 5.1% |
| Social media ads | 2.9% |
| Bing search ads | 1.9% |
| Banner ads on other sites | 0.3% |
Say 900 visitors come from banner ads and 100 from your email list. At those rates, you'd get about 3 sign-ups from the ads and 17 from email, so your page shows 2%. It looks like a bad page. Your real issue is where 90% of your visitors come from.
Be patient with small numbers. At 200 visitors, 3% is six people and 4% is eight people. That jump is two humans deciding something on a Tuesday. Until you have at least a few hundred visitors from one source, treat your rate as a rumor.
A higher rate can still mean fewer customers
This part affects your bank account. Here are First Page Sage's numbers for paid visitors, following 10,000 of them all the way through:
| Type of offer | Visitors who sign up | Who later pay | Paying customers |
|---|---|---|---|
| Free trial, no card | 7.1% (710) | 17.4% | About 124 |
| Free trial, card required | 2.2% (220) | 51% | About 112 |
| Free account that never expires | 15.9% (1,590) | 2.8% | About 45 |
These are averages from different companies, so treat this as a sketch. It still makes the point. The page that asks for a card looks three times worse than the no-card page, but ends up close to it. The free account gets the most sign-ups and the fewest paying customers in this table. Long-time free users do sometimes upgrade as their needs grow, so give it some patience.
Two things follow. If you're newer, start with no card, because you need people in the product and talking to you. And check how long people stay before you celebrate a card-required trial, because that 51% counts anyone who paid for a single month.
The same goes for demos. A page that doubles its requests by attracting people who can't buy has only made your calendar worse. Save where every lead came from in your customer database. Then count, for each page, how many booked calls happened and how many became customers.
Find what's actually wrong
| What you see | What it probably means | Start with |
|---|---|---|
| Fine from email, poor from ads | The ad and the page don't match, or the visitors have never heard of you | Your visitors |
| Poor from every source | People can't tell what it is or who it's for | The first screen |
| Lots start the form, few finish | The form is too long, or they fear what comes next | The form |
| Healthy rate, few customers | The wrong people are signing up, or the trial isn't showing them why it's worth paying for | The section above |
| Under a few hundred visitors | You don't have a landing page problem yet | Talking to people |
For the form row, it helps to count people who start the form as well as people who finish it. About half finishing is typical. If you're near that, shortening the form won't help much, and the problem sits higher up the page.
Visitors come first. Read your ad, then read your headline. If the ad promises "automatic invoice reminders" and the page opens with "the all-in-one finance platform," the visitor has to do the translating, and most won't bother. Give each major ad its own page or at least its own headline, using the same words.
How to fix a free trial page
Say what it does in words a tired person understands. In the same Unbounce data, pages written at a 5th to 7th grade reading level got 12.9% of visitors to sign up. Pages written at a professional level got 2.1%. So:
- Before: "The AI-native workflow orchestration platform for modern finance teams"
- After: "Stop chasing invoices. Acme reminds clients until they pay."
Your headline is your value proposition in one line: what you do for someone and why it matters. Reading a few value proposition examples will quickly show you how plain the good ones are. Show the first screen to a friend for five seconds. If they can't say what it does and who it's for, rewrite it before you touch anything else.
Answer the questions sitting under the button. Put one short line there: "14 days free. No card needed. You'll be set up in about two minutes." Only use the bits that are true.
Ask for as little as you can. An email and a password, or a sign-in with Google, is enough to begin. Ask about role and team size once they're inside.
Show the real product. A screenshot or short clip of the actual screen beats a drawing of people high-fiving near a laptop. If setup takes a while, a clickable tour lets people see it first. FORM, a task management tool, reported 33% more sign-ups on its paid landing pages after adding tours. That's from the tour company's own case study, so your results may differ.
Put proof beside the button. A named customer saying what changed is worth more than a row of logos no one recognizes.
Think about phones. Unbounce found 79% of visits to software landing pages were on phones. If your product really needs a computer, a trial started on a phone goes nowhere. Offer a second button: "Email me a link for later." You keep the person, and they come back when they can actually use it.
How to fix a demo page
A demo request costs the visitor far more than a trial does, because a person will be contacting them. Most of the fixes here are about making that feel safe.
Send it people who already know you. Don't treat the demo page as a first stop. People who've looked at your pricing, searched your name or read a comparison are ready. Anyone who's just met you belongs on a product overview or your saas homepage first.
Say exactly what happens next. Many people hesitate here because they're picturing a week of sales calls. Tell them something like this: "We'll reply within one working day. You'll get 20 minutes with Sam, who'll ask how your team handles this now and show you the parts that matter. If it isn't a fit, we'll say so." Write your own version, and make sure sales will actually do it.
Name the button honestly. "Request a demo" suggests they'll see the product, but often they get a call to check whether they can buy. "Book a 20-minute call" or "See if it's a fit" sets the right expectation.
Offer a smaller step for people who aren't ready. Some visitors want to look before they talk. In the FORM case above, a "Take a tour" button cut the cost of each lead by nearly half, compared with "Request a demo." That was in campaigns aimed at specific companies. A recorded demo does a similar job for less work.
Cut the form to what sales uses. Every extra box gives people another reason to stop. Open your last 20 leads and see which boxes anyone looked at. Delete the rest. Keep the one question that best tells you someone can buy, such as company size.
Show a starting price. It feels risky, but it filters out people who were never going to buy and saves everyone a wasted call. You don't need the full price table. Some pricing page examples will give you ideas for a short version, and "Plans start at $X a month" is often enough.
Let people who know you book straight away. If most visitors already know you, put a calendar after a short form. If they mostly don't, keep the screening step, so your calendar doesn't fill with students and competitors.
Trim the menu, but leave a way out. Don't strip out every link. A few links that jump down the page and a small footer let someone who isn't ready read more without hitting the back button.
Skip the A/B test and ask people
Testing two versions of a page (an A/B test) takes more visitors than most small companies have. If 3 of every 100 visitors sign up and you want to spot a 20% improvement, a standard calculator says you need about 13,000 visitors per version. Most small companies would wait months, just to learn about one change.
So make bigger, bolder changes one at a time, and compare before and after from the same source of visitors. Keep testing for later, when conversion rate optimization becomes a steady habit and you have the visitors to support it.
In the meantime, ask. Put one question on the page and show it after someone has been there a while:
What's stopping you from signing up today?
Offer tappable answers, then act on them:
| They say | What to do |
|---|---|
| I'm not sure it does what I need | Put that use case on the first screen, and add it to your FAQ |
| I need to see the price | Add a starting price |
| I don't want a sales call | Say what happens next, and offer a tour |
| I need to check with my team | Give them something easy to forward |
| I'm just researching | Offer a way to come back, like a recorded demo |
| Something else | Read these first. They're the reasons you didn't think of. |
Then email your last ten sign-ups and ask what almost stopped them. You'll learn more from three honest replies than from a month of staring at a dashboard.
A survey tells you why someone's hesitating, but they still leave. If people keep saying "I'm not sure it does what I need," they needed an answer while they were on the page. That's what we built Possible to do, so take this with the usual pinch of salt: it's our own product. Possible is an AI built on real sales psychology. It asks each visitor the right question at the right moment, remembers the answers and guides them to a sign-up or booking. It's free to start, and there's more on our AI sales assistant for SaaS. If your visitors are wrong or your headline is unclear, fix those first.
The one-hour version
- Look up your rate for your main page, split by where visitors came from.
- Compare it with the right row at the top of this page, for the right source.
- Fill in your own form on your phone. Check that it works and gets counted.
- Rewrite your headline so a tired stranger understands it.
- Add one line about what happens after they click.
Questions about landing page conversion rate
What is a good conversion rate on a landing page?
It depends on what you're asking for. Across all industries, the typical page in Unbounce's data gets 6.6%. For software, about 4% to 8% is normal on a no-card trial page and 1% to 4% on a demo page.
Is a 12% conversion rate good?
Usually, yes. On a demo page with paid visitors it's excellent, and it's well above normal for a no-card trial. Check the source, though. The typical software page gets 16.9% from email, so 12% from your own list may be nothing special.
Is 2.5% a good conversion rate?
On a demo page, yes. On a trial that requires a card, it's about typical. On a no-card trial it's low, so look at where your visitors come from and at your first screen.
What is a typical conversion rate for SaaS trials?
There are two numbers. Of visitors, about 4% to 8% start a trial with no card, and 2% to 2.5% with one. Of people on a trial, about 12% to 18% go on to pay with no card, and about 50% with one. These come from industry reports, so treat them as rough.
How do you calculate landing page conversion rate?
Divide sign-ups by visitors and multiply by 100. So 40 demo requests from 1,000 visitors is 4%. Work it out for one page, one action and one source of visitors at a time.
Should I require a credit card for my free trial?
If you're early, probably not. You need people using the product and giving you feedback. A card brings fewer sign-ups and a higher share who pay, so the customer count can end up close. Compare paying customers per visitor, and check how long they stay.
How much should a landing page cost?
There's no standard price. It can be a cheap template or a specialist's fee. Before you pay for a new page, check where your visitors come from and whether your tracking works, because a redesign fixes neither. Then ask whoever you hire how they'll measure success, and whether they'll count customers or only filled-in forms.
Go deeper on conversion tactics.