On this page
What actually differs
An instant form, sometimes still called a lead form, opens inside Facebook or Instagram when somebody taps your advertisement. Meta pre-fills the name, email and phone number it already holds, the person confirms, and the lead arrives in your system. They never leave the app and the whole interaction can take four seconds.
A landing page sends them to your website, where they read something, look at photographs, form an opinion and then either ring, fill in a form or leave. It takes longer, more people abandon, and the ones who do not have invested something.
Everything that follows is a consequence of that difference in effort. Effort is not friction to be eliminated. In lead generation, effort is information: it tells you something about the person that no targeting setting can.
Why forms produce cheaper leads
Three mechanical reasons, none of them mysterious.
First, the drop-off between tapping an advertisement and completing an action is where most leads are lost, and the instant form removes almost all of it. There is no page to load, no scrolling, no typing.
Second, the form removes your website from the equation. If your site is slow, dated, or built for desktop and being viewed on a phone on a bus, that weakness disappears. For a trade firm whose site has not been touched since 2019, the instant form is often doing a great deal of quiet work covering for it, which is worth knowing before you conclude that forms simply perform better.
Third, Meta optimises towards whichever event you tell it to value. Give it a cheap, easily-achieved event and it will find people who complete cheap, easily-achieved events. That is the system working exactly as designed, and it is also the root of the problem in the next section.
The cost gap is real and routinely substantial. It is not unusual for an instant form campaign to produce leads at a third or half the price of the same advertisement pointed at a website. If cost per lead were the goal, the argument would be over.
Why those leads are worse
Cost per lead is not the goal. Cost per job won is the goal, and the two diverge sharply here.
What actually arrives from an unqualified instant form campaign includes a substantial proportion of people who do not answer the phone, several who do not remember enquiring, some who are outside your working area, some who wanted a different service entirely, and a handful of mis-taps from people scrolling quickly. The pre-filled details mean a person can submit without reading a word of the form, and some of them do.
There is a second, subtler problem. Because the person never visited your website, they arrive at the phone call knowing nothing about you. They have not seen your work, your guarantee, your reviews or your face. Every bit of persuasion your site would have done now has to be done live by whoever picks up, on a call the prospect was not expecting. Firms with a strong website are effectively switching it off when they use forms, and then wondering why the conversations are harder.
None of this makes instant forms wrong. It makes the raw cost-per-lead comparison meaningless. A campaign producing leads at half the price that converts at a quarter of the rate is a worse campaign wearing a better number.
What the platform learns from your choice
There is a consequence to this decision that most firms never see, and it outlasts any individual campaign. Meta's delivery system is not neutral about what you ask it to optimise towards. It builds a model of who completes your chosen event and then goes looking for more people resembling them.
Point it at an in-app form submission and it learns the characteristics of people who tap forms readily. Those people exist in quantity, they are cheap to reach, and a fair number of them tap a great many forms belonging to a great many firms. Over a few weeks the audience drifts in that direction and the account settles into a pattern that is awkward to reverse without starting the learning over.
Point it at a deeper event and it learns something more useful. The deepest signal available to a trade business is not a form submission at all: it is the job you actually won, fed back to Meta as an offline conversion through the Conversions API once your own system has recorded it. That is a slower loop, it requires plumbing that does not exist by default, and it is the closest a small firm gets to telling the delivery system what a good customer looks like in its own business rather than in general.
This matters more than it used to, because browser-only tracking is leakier than it was. Declined app tracking on iOS, restricted third-party cookies and ad blockers all remove events from the pixel's view, and a model degrades with the data behind it. A server-side feed rebuilds part of what was lost. The practical upshot for the choice in front of you: if your tracking is thin, instant forms will look better than they are, because an in-app submission is one of the few events the platform can still count reliably.
Making instant forms behave
If you are going to use them, and for many trade campaigns you should, build them to cost you volume deliberately.
- Use the higher-intent format. Meta offers a version that adds a review step before submission, requiring the person to confirm rather than tap straight through. It reduces volume and improves what arrives, which is the trade you want.
- Add custom questions. Two or three, answerable in a tap. Property type, rough size or scale, and timescale. Avoid free text, which people abandon.
- Ask something only a real prospect knows. Approximate square metres, number of windows, the age of the boiler. It takes a second for a genuine enquirer and stops an idle one.
- Use conditional logic where the service branches. A different follow-up question depending on the first answer both qualifies and signals seriousness.
- Say the unglamorous part out loud. A line stating that you cover a specific area, or that there is a minimum job size, costs you the enquiries you did not want.
- Overwrite the pre-fill on phone number. Where the option exists, requiring the number to be typed removes a meaningful share of the accidental submissions.
Expect your cost per lead to rise, possibly sharply. Judge the change on contactable, qualified enquiries rather than raw count, and give it long enough to see the difference in booked work rather than in the dashboard.
When a landing page wins outright
Four situations where sending people to your site is clearly correct.
High-value, considered purchases. Anything where the customer needs to see evidence before they will speak to you. Kitchens, extensions, glazing, solar, full re-roofs. These buyers want photographs, price context, process and proof, and an instant form denies them all of it at the moment they were willing to look.
Warm and retargeting audiences. Somebody who already visited your site, watched a video or engaged with the page has demonstrated interest. Sending them to a page that builds on it is more productive than extracting a phone number from them as quickly as possible.
When the offer needs explaining. A paid design appointment, a survey with a fee, anything conditional. The explanation does not fit in a form and an unexplained offer converts badly on the call.
When your site is genuinely good. If you have a fast, phone-first site with real photography of your own work and clear pricing context, it will out-convert an instant form on quality of enquiry almost every time. Sending traffic away from a strong site to a generic in-app form is giving away the advantage you paid to build.
The variable that beats both
Neither mechanism matters as much as how fast you respond, and this is the single most common failure in trade lead generation.
A Meta lead is at its warmest in the first few minutes, because the person has just seen the advertisement and remembers why they tapped. An hour later they are back at work. Tomorrow they have forgotten. Instant form leads decay faster than any other lead type precisely because the interaction was so brief, and a firm that batches its callbacks to the end of the day is destroying most of the value before it picks up the phone.
Practical minimum: an automatic text or email within sixty seconds confirming a real person has it and when they will ring, then an actual call the same morning or afternoon. Leads sitting in a Facebook forms library that nobody downloads for a week are worse than no leads at all, because you paid for them. Pipe them straight into your CRM with a notification attached. This is dull plumbing and it changes results more than any change to the advertisement.
Choosing, campaign by campaign
Do not pick one for the account. Pick per campaign, based on what the audience has already done.
- Cold prospecting, low-ticket or impulse-adjacent work. Instant form with qualifying questions. Gutter clearing, moss treatment, hedge work, small repairs.
- Cold prospecting, high-ticket considered work. Landing page. The evidence has to do the selling before a conversation is worth having.
- Neighbour and proximity campaigns. Instant form usually wins. Interest is fresh and specific, the ask is small, and the local relevance has already done the persuading.
- All warm and retargeting layers. Landing page. They have shown interest; give them something to act on.
- Recruitment. Instant form, every time. Nobody completes a careers page on a phone at half past nine in the evening.
Then measure the thing that decides it. Not leads, not cost per lead, but contactable enquiries, quotes issued and jobs won, split by mechanism. Most trade firms discover the split is not what they assumed, and a few discover the expensive mechanism is the cheaper one once the work is counted. If you want to model that before committing, the cost per lead calculator lets you run both conversion rates side by side, and the benchmarks page explains why we will not hand you an industry average to substitute for your own numbers.
Rather have someone else do this?We run exactly this for trade and construction firms across the UK. Call and we will tell you what is realistic in your area, including when the answer is that it is not worth it.
Call 07443 392243WhatsAppCommon questions
Will adding qualifying questions ruin our lead volume?
It will reduce it, deliberately, and that is the point. Expect fewer leads at a higher cost per lead. What you are buying is a higher proportion who answer the phone, are in your area and actually want the job. Judge the change on quotes issued rather than leads received, and give it at least six weeks so the comparison covers the same kind of month.
Can we run both at once and compare?
Yes, and it is the sensible way to settle the argument for your own business. Run the same creative and audience with the two mechanisms as separate ad sets, give each enough budget to gather real volume, and compare after a few weeks on jobs won rather than leads received. Make sure your records distinguish the two sources or the test tells you nothing.
Why do our lead form enquiries never answer the phone?
Usually a combination of three things: the submission took almost no effort, the details were pre-filled so the number may be an old one, and the callback came hours or days later. Fix the last one first because it is free. Then add the review step and qualifying questions. If the problem persists, the audience is too broad or the creative is attracting curiosity rather than intent.
Does the landing page need to be a special page?
It needs to match the advertisement and load fast on a phone. A dedicated page for the specific service, with your own photographs, honest price context and a phone number that is tappable above the fold, beats sending everyone to a homepage by a wide margin. It does not need to be a separate microsite, and in most cases a good service page on your existing site is enough.