Cut Meta for two weeks, lose 80% of your leads

We ran a test on a project this year that’s worth sharing again because it keeps coming up in client conversations: turn off Meta Ads for two weeks, kept only one of the platforms on (the red one), running as normal.

Result: an 80% drop in lead volume. No improvement in lead quality. A project typically pulling around 30 leads a month across portals dropped to two leads in the two week window.

The takeaway isn’t that portals don’t work. It’s that portals are often harvesting demand that’s been created somewhere else, Meta, Google, hoarding, letterbox drops, and rarely generate much on their own. Switch off the channel doing the upper funnel work and the lower funnel channels have nothing left to catch.

It’s also why we’re seeing more clients try to solve a Meta quality concern by shifting budget into portal add ons instead. More on why that doesn’t work below.

We’re seeing more clients ask to pull back on Meta when lead quality is the concern, and shift that budget into portal add ons like REA’s Audience Extension instead.

Worth knowing before you make that call: Audience Extension is largely served through the same Meta ad network, using overlapping audience signals. In practice, you’re not diversifying your reach, you’re paying twice to show up in front of the same person.

If the issue really is quality from Meta, moving spend into a product that targets the same people won’t fix it. The lever that actually moves quality sits elsewhere: the form, the landing page and the qualifying questions you ask before someone submits, not which platform served the ad.

Our recommendation: before shifting a dollar from Meta into Audience Extension, ask REA (or whoever’s selling the placement) to confirm how much audience overlap exists with your existing Meta campaign. If it’s high (which it is), you’re solving a quality problem with a volume solution.

The solution moves away from channel and to; Creative and most importantly is the Price of the project what people are willing to pay.

Spring is usually the season property campaigns pick up pace, but this year’s likely to run a little differently. With the federal election set for November, we’re expecting buyer hesitancy to stretch longer than a typical spring cycle, and that pattern may not properly lift until after the result is known.

What we’re seeing in the data isn’t disengagement. Leads are still coming in and buyers are still enquiring, they’re just taking longer to move from interested to committed. People are spending more time researching before they’re ready to commit.

That shift puts more weight on the nurture sequence than a normal spring. If your campaign is built to expect a fast decision the moment someone shows interest, a slower spring will look like a drop off when it’s actually a longer research window. Worth reviewing your nurture sequence end to end, first touch through to sales handover, and checking it’s giving people enough reason to stay warm over a longer runway rather than one built for a faster cycle.

What does that mean: Ensuring you have different touch points from initial interest (top of funnel) through to education (middle of funnel) and finally harder tactical sales push (bottom of the funnel).

There’s a lot of noise right now about AI answering the phone, chatbots doing instant callbacks, voice assistants qualifying buyers before a human ever gets involved. It’s real technology and it exists. But we’re not seeing it land at scale in Australian property yet. We’re laggards on voice search generally, and there’s still a trust gap, most buyers would rather type into a form than talk to a bot.

What we are seeing work is less flashy and more useful: AI doing the grunt work behind a campaign rather than in front of it. That means using AI to help build sharper qualifying questions into the form and campaign structure itself, so the leads that come through have already self-selected on budget, timeline and intent, rather than a form that just captures a name and email.

The result isn’t more leads. It’s fewer, better qualified ones. Sales teams end up spending time on genuine, warm conversations with people who are ready to talk, instead of working through a pile of enquiries where half were never going to convert.

If you’re considering where AI actually earns its place in your property marketing right now, it’s in the qualifying layer, not the phone line (yet, but that will happen soon).

There’s a limit.

Good funnel marketing on Meta means a small number of ads doing distinct jobs, top of funnel awareness creative introducing the project, mid funnel creative building consideration (construction updates, suburb content), and a tighter retargeting set doing the actual conversion work. Flooding the account with variations of the same message doesn’t build a funnel, it just adds noise and eats budget testing near identical creative against itself.

Our rule of thumb: fewer, sharper ads outperform a large volume of similar ones. Somewhere between 2 and 5 live ads per campaign stage is usually the sweet spot, enough to test and avoid fatigue, not so many that spend gets spread too thin to learn anything.

What we recommend: review and refresh your live ad set every 6 weeks. Not a full rebuild, just pull what’s underperforming, keep what’s working, and add one or two new variations to test. That cadence keeps creative fresh without falling into the crowding problem we’re seeing elsewhere in the market.

• Don’t turn off Meta, it’s the only channel driving real leads and more importantly, it’s the most affordable.
• Check audience overlap before shifting Meta budget into Audience Extension (or similar portal add ons). If overlap is high (which it is), it won’t fix a quality problem, it’ll just double up spend on the same people. You’ll spend more money to find the same people.
• Use AI to sharpen qualifying questions in your forms and campaigns, not to chase a phone answering gimmick our Australian buyers aren’t ready for yet.
• Cap live ads at 2 to 5 per campaign stage, and refresh the set every 6 weeks rather than adding to the pile.

As always, happy to share what we’re seeing across the market. No pitch, just data.