Active demand, the right bucket and a pressure-tested channel. Walk into any audit of an underperforming SA paid media account and you’ll find the same pattern: The founder likely spread R20,000 across Google, Meta, LinkedIn and TikTok, hoping something would stick.

None of those had enough budget to learn, and now no one knows if it was the right channel. The fix usually isn’t “spend more.” It’s spend on the right thing first” And the question that decides what right means is all about whether your customers know they need you yet.

Chantelle Bowyer’s been doing this for almost 16 years. She runs Metis, a paid media agency that audits SA accounts and routinely finds extreme amounts of wasted spend. She used to be a Google trainer back when in-person training was still a thing. So she’s been on both sides: building the playbook for Google, and then watching SA businesses break it in predictable ways.

Here’s the framework she uses to figure out where your first rand should go.

Every paid media decision starts with one question: Are you trying to be found by people who are already looking, or are you trying to convince people who aren’t looking yet?

Those are completely different jobs. They use different channels, creative, metrics and conversion expectations. “Capturing demand is roughly 4 to 6% of the market — people with a need going online to find an answer. Generating demand is everyone else. People don’t realise this solution exists, or don’t realise they have the pain point.”

That single split decides almost everything else.

Capturing demand only works if demand exists. The test is simple: Would a person with this problem actually type something into Google to find a solution?

A plumber, an accountant, a payroll system, a CRM, a divorce lawyer — all yes. Nobody calls a plumber unless something’s leaking. They go to Google, type in emergency plumber Cape Town, and pick from whoever shows up first. That’s captured demand. The intent is already there. Your job is to be findable.

A new productivity app, a niche subscription box, a new category of fintech, a workshop nobody’s heard of — mostly no. So you have to generate the demand.

Chantelle’s rule: chase the existing demand first. It converts faster, it’s cheaper and it teaches the platform who your buyer actually is. “I always go after capturing demand first until that’s exhausted, then look at generating demand. Some businesses have to start with generating demand, depending on what they sell.”

Once you know which bucket you’re in, the channel choice mostly makes itself. Capturing demand lives on Google Search, where intent shows up as a query. Warmer leads, faster conversions, better economics on the first sale.

Generating demand lives on social and video. Meta, YouTube, LinkedIn — places where people are scrolling, not searching. The job is to introduce the problem, the solution and your brand to people who weren’t looking. 

That requires educational creative, not a search ad. If you sell something with active search demand and you’re only running Meta ads, you’re leaving the cheaper, warmer channel on the table. If you sell something nobody’s searching for and you’re only running a Google search ad, you’re bidding against air.

Founders default to whichever channel feels familiar. The audit reveals the truth. If you’re convinced you should be running Google ads, open Google Keyword Planner and look up the actual search volume for the terms a customer would use. Not your brand. Not your product name. The problem they have in their language. 

If the volume is real, capturing demand is on the table. If there are 30 searches a month nationally, the search isn’t going to scale and you need to be generating demand instead.

If you’re convinced you should be running Meta or LinkedIn ads, ask yourself a different question: Do you have the creative budget and the educational content to actually convince a stranger? Generating demand is a longer game with weaker conversion at the click level. The metric that matters isn’t “did they buy today” — it’s “did they enter the funnel.” If you’re going to measure a generate-demand campaign by purchase conversions in week one, you’ll kill it before it has a chance to work.

Be honest about which channel your business actually needs, not which one you saw your competitor running.

Once you know which side you’re on, do not split a small budget across both. This is the single most common mistake in SA paid media accounts.

The platforms need data to learn. The rule of thumb Chantelle uses: 30 conversions per campaign per month is the floor for the algorithm to actually optimise. Below that, the campaign sits in learning mode forever and never gets out of the gate. R20,000 split four ways is four campaigns getting nowhere. 

“R20k budget split into R5k Google, R5k LinkedIn, R5k Facebook, R5k TikTok — none of those are big enough to learn.”

Pick the bucket. Pick the channel. Put real money behind it. Add the second channel only when the first one is dialled in.

Capturing demand campaigns get judged on cost per acquisition, return on ad spend and conversion rate. The intent is high, so the bar is high. If your search ads aren’t producing sales or qualified leads inside the first month or two, something is wrong with the targeting, the landing page, or the offer.

When generating demand, campaigns get judged on engagement, video view rates, click-throughs, and movement into the funnel. Sales come later, often through a remarketing campaign that catches the same person back on search after the educational ad introduced them to the category. Hold these campaigns to the wrong metric and you’ll switch them off the day before they would have started working.

Two different jobs. Two different scoreboards. Don’t mix them.

First, SA paid media budgets are small and most founders here aren’t spending what a US business spends in a week. That makes the cost of spreading thin much higher here than overseas — you don’t have the buffer to fund four campaigns until one figures itself out. Concentration is how SA accounts get to performance.

Second, Google’s defaults aren’t built for the SA market. Their best practices come from large US budgets and broad-match assumptions that don’t translate to a smaller, more fragmented audience. SA accounts that work tend to start tighter, with exact-match keywords and clearly defined audiences, then broaden once data is flowing.

“Google still has 5x more traffic than ChatGPT as of two months ago. They do not think of South Africa when putting best practices together.”

When you pick the right bucket first, paid media gets easier: The channel choice is obvious, the budget is concentrated where it can actually learn and your metrics line up with what the campaign is supposed to do.

It takes a single afternoon to make the call. It saves you the months of wasted spend that come from getting it wrong.

This workflow is one piece of Paid Ads 101, Chantelle’s full masterclass inside the Founder Collab. The full session goes in depth on the audit findings she most often sees in SA accounts and the systems she uses to fix them. 

Inside the full masterclass:

You’ll also get access to 40+ other masterclasses from SA founders and operators on sales, fundraising, UX, automations and more inside The Founder Collab.

This workflow first appeared in our 5 May ‘26 edition, featuring the Matric Live free EdTech app.

Get more SA tech and business news and subscribe to The Open Letter.