How do you set up goal tracking for a demo request form?

How do you tell whether anyone is actually submitting your demo form?

Give the successful submission its own URL, then mark the resulting event as a key event in Google Analytics. That is the whole job. Most sites skip the first half, which is why their analytics show traffic to the demo page and nothing at all about whether the form worked.

This is the single most common measurement gap I find. The form is fine, the notifications arrive, the sales team gets the leads. Nobody can answer the question "did the new homepage increase demo requests", because the submission was never recorded anywhere a report can reach.

What follows is the sequence I use, written for one situation: you have a demo request form on a Webflow site, you have GA4 installed, and you want a number you can trust by tomorrow.

Why does your form not show up as a conversion already?

Because nothing changed that analytics could see. A form that swaps in a thank-you message on the same page leaves the URL untouched, and page-based measurement has no way to tell a successful submission from a visitor reading the page. No new pageview, no event, no conversion.

Installing analytics gives you pageviews automatically. It does not give you an opinion about which visitor actions matter, and form submission is an action rather than a page. Somebody has to declare it, and that somebody is you.

So the gap is not a broken tag. It is a missing definition. Until you tell the system that a particular thing counts, it will faithfully report everything except the thing you care about, which feels like a bug and is actually a default.

What should you count as the conversion, the submit or the thank you?

The thank you, always. Counting button clicks counts intentions, including the ones that failed validation, hit a server error or were the same impatient person clicking twice. Counting the thank-you state counts submissions that actually completed, which is the number you will quote to someone.

The gap between the two is not small, and it is the interesting part. If clicks are far above completions, you have a form problem rather than a traffic problem, and no amount of extra visitors will fix it. Measuring only clicks hides exactly the failure you most need to see.

There is an honest cost to the thank-you approach: you lose visibility into abandonment unless you measure both. My preference is to make completions the official number and treat click data as diagnostic. One number for reporting, one for debugging, and never confuse which is which. While you are in there, it is worth asking whether the form is asking for the right things in the first place.

How do you give the submission its own URL in Webflow?

Build a dedicated thank-you page, then set the form to send people there after a successful submission rather than showing a message in place. In Webflow that behaviour lives in the form's settings panel, and I would check Webflow's own Help Center for the exact current steps because the interface labels move.

Make the thank-you page a real page with a real path, something like /demo-thank-you. Give it useful content too, because a visitor who just asked for a demo is the most engaged person on your site and a dead end is a waste of that moment. Tell them what happens next and how long it will take.

One caution that costs people weeks: keep that page out of your sitemap and out of search results. A thank-you page that gets indexed will collect organic visitors who never filled in your form, and every one of them will be counted as a conversion. Your number becomes fiction, slowly, in a way nobody notices for a quarter.

How do you turn that into a key event in GA4?

Once the thank-you page has its own URL, you create or identify the event that measures that view, then mark that event as a key event. Google's documentation describes key events as measuring actions that are important to your business and letting you evaluate the marketing channels that lead users to take those actions.

The marking step is the part worth writing down. In Google Analytics you go to Admin and then Events, and Google's documentation says to mark the event as a key event by clicking the star icon next to it. Interface wording in this product changes often enough that I would read the current help page rather than trust a screenshot from anyone, including me.

You also need the right permission. Google's documentation states you must be a marketer or above at property level to mark events as key events. If the star does not appear for you, that is usually the reason rather than a missing event, and the fix is an access request rather than an hour of debugging.

How long before the number appears in your reports?

Not immediately, and this catches almost everyone. Google's documentation says to allow for up to 24 hours for a key event to show up in standard reports, while realtime reports update a few minutes after a user triggers a key event. Those are two different clocks.

So the correct way to test your setup is to submit the form yourself and watch the realtime report. If it appears there within a few minutes, the plumbing works. If it does not, something is genuinely wrong and you can start looking now rather than tomorrow.

What you must not do is check the standard report an hour later, see zero, and start rebuilding a configuration that was correct. I have watched people undo a working setup because they tested it against the wrong clock. Verify in realtime, report from standard, and wait the day before you conclude anything.

How many key events are you allowed to have?

Fewer than you will want, which is a useful constraint. Google's documentation states you can mark up to 30 events as key events for standard properties and 50 events as key events for Analytics 360 properties. That is a ceiling worth knowing before you start marking everything interesting.

In practice this is a feature. A property with 30 key events has no key events, because nobody can hold thirty priorities. The discipline I apply is that a key event should be something you would mention in a board update or a client report, which usually leaves you with three to five.

Demo requests, trial starts, pricing page contact and maybe newsletter signup. Scroll depth and video plays are engagement signals, not key events, however satisfying they look on a dashboard. Keep the ceiling in mind and spend it deliberately rather than discovering it later.

What will make this number wrong?

Four things, and all of them are quiet. An indexed thank-you page collecting organic traffic. A reload of the thank-you page counting twice. Your own team submitting test forms into production. And a second tool counting the same thing differently, so two dashboards disagree and neither is trusted.

The reload problem is worth guarding against specifically, because a visitor who refreshes or returns to that URL later will trigger the event again. Whether that matters depends on your volume, but you should know it is happening rather than discover it when a month looks unusually good for no reason.

Internal testing is the one I am strictest about. Once the setup works, stop submitting real test entries, or filter your own traffic, or use a staging form. Otherwise your conversion count includes your own curiosity. When two systems then report different totals, that is a familiar problem, and I have written about what to do when two tools disagree about a number.

What should you do next?

Do it today and verify it in the realtime report before you close the tab. Build the thank-you page, point the form at it, keep it out of your sitemap, mark the event as a key event, then submit the form yourself and watch it arrive. Under an hour of work, start to finish.

Then wait a day and look at the standard report, because that is the number you will actually use. Once it is running, resist the urge to add more key events. The value here is having one trustworthy number, not a dashboard with everything on it. Pair it with campaign tracking and you can finally attribute demos to sources, which is where UTM tracking for Webflow campaigns earns its keep.

Across 70+ projects for 25+ clients over 6+ years, this is the setup I have rebuilt more than any other, almost always because nobody did it at launch and nobody could prove anything afterwards. If you are about to redesign a page and cannot currently measure whether it worked, reach out and let's get the measurement in before the redesign, not after.