Most Google Ads copy fails the same test: it describes the product instead of interrupting the scroll. If your CTR sits under 3% on search campaigns in 2026, the fix usually isn't the bid or the keyword list. It's the fifteen words sitting in your headline.
TL;DR
Google ads copy formulas that lift CTR share three traits: a specific number, a stated outcome, and a reason to click now instead of later. Ads built on vague benefit statements ("quality service", "trusted provider") consistently underperform ads that name a price, a timeframe, or a result. Rebuilding headlines around one formula per ad group, tested against a control, is the fastest lever available to South African advertisers running Google Ads management for Johannesburg businesses or any other local account in 2026. Verdict: rewrite before you rebid.
Why this matters
CTR isn't a vanity metric in Google Ads. It's a direct input into Quality Score, and Quality Score is a direct input into what you pay per click. An account with a 2% CTR and one with a 6% CTR on the same keyword can pay wildly different amounts for the same position, sometimes 30 to 50% apart depending on competition density.
Most accounts we've audited across Johannesburg and Cape Town in 2026 have the budget and targeting basics right. The copy is where the money leaks. Generic headlines get generic prices. Specific headlines get clicked.
What you'll need
- Access to your Google Ads account with edit rights on at least one active campaign
- Search terms report from the last 30 to 90 days
- Three to five real numbers about your business: price, delivery time, stock levels, guarantee period, or customer count
- A competitor's top three ads (search your main keyword and screenshot what shows above you)
- Two hours, split across a planning session and a copy-writing session
- Access to Google Ads Editor if you're rewriting more than five ad groups at once
The steps
1. Pull your current CTR by ad group, not by campaign
Campaign-level CTR hides which ad groups are actually the problem. Open the Ads & Assets tab, segment by ad group, and sort by impressions descending. Anything with over 500 impressions and under 3% CTR in the last 30 to 90 days is your priority list for 2026.
Common mistake: rewriting the whole account at once. Fix the five worst ad groups first, measure for two weeks, then move on.
2. Identify which formula your current headline is using (if any)
Most underperforming headlines fall into one of three lazy patterns: brand name plus generic descriptor ("ABC Plumbers – Quality Service"), keyword stuffed with no benefit ("Plumber Johannesburg Plumbing Services"), or a vague claim with no proof ("Trusted Experts You Can Rely On"). None of these give a reader a reason to pick you over the three other ads on the page.
Write down which pattern each underperforming headline matches. This tells you what to replace it with.
3. Rebuild headlines using the number-plus-outcome formula
The strongest performing formula across most South African search accounts is: [specific number] + [outcome] + [timeframe or condition]. "Geyser Replaced in 4 Hours" beats "Geyser Repair Services" because it answers the two questions a searcher actually has: how fast, and what happens.
Apply this to at least three headline slots per ad. Google rotates combinations, so give it real material to test, not three versions of the same sentence.
4. Add a friction-removal line in Headline 2 or 3
Once the outcome headline is set, use a second headline to remove the objection that stops someone clicking. That's usually price, commitment, or trust. "No Callout Fee" or "Quote in 24 Hours" does more work than "Family Owned Since 1998" because it answers a buying question rather than a brand question.
This matters more for high-consideration categories. A Google Ads for real estate agents campaign, for instance, gets more clicks from "Free Property Valuation This Week" than from "Award-Winning Agency", because the first line answers what happens if you click.
5. Use description lines to stack proof, not repeat the headline
Descriptions are wasted space if they just restate the headline in longer form. Use them to add a second number: stock count, review volume, response time, or coverage area. "Same-day quotes across Gauteng. Rated 4.8 from 200+ reviews" gives the reader two separate reasons to trust the click, not one reason said twice.
6. Match ad copy to the actual search term, not the broad theme
Pull your search terms report and check whether your ad copy speaks to the specific intent behind the top 10 terms by volume. An ad group targeting "emergency plumber Sandton" needs urgency language. An ad group targeting "plumbing quote Sandton" needs a price or process line. Running the same three headlines across both wastes the intent signal you already have for free.
7. Set up a genuine A/B test, not a rotation-and-hope
Pause the old ad only after the new one has 1,000+ impressions and at least 2 weeks of data. Ending a test early on small sample sizes is the single most common reason advertisers think a formula "doesn't work" for their category, when actually the sample never reached significance. Google's own asset performance labels (Best, Good, Low) are a start, but they're not a substitute for checking raw CTR yourself.
8. Log every winning formula in a swipe file
Once a headline formula beats a control by 20% or more in CTR, write it down with the ad group, the date, and the percentage lift. By the third quarter of testing you'll have a formula library specific to your business, which is worth more than any generic "ad copy templates" list because it's calibrated to your actual buyers.
Troubleshooting
CTR improved but conversion rate dropped. You're attracting clicks with a promise the landing page doesn't deliver. Check that the number in your headline (price, timeframe, stock) appears on the landing page above the fold too.
New headline has high CTR but Quality Score didn't move. Quality Score updates lag actual performance by roughly one to two weeks in most accounts. Give it time before concluding the formula failed.
Ad strength shows "Poor" despite good CTR. Ad strength is a diversity score, not a performance score. Add more unique headline variations rather than rewriting the ones already working.
Competitor ad copy looks identical to yours. If three advertisers on the same page use the same generic claim ("trusted", "quality", "professional"), that's the opportunity. Whoever adds the first real number wins the click by default.
CTR is high but cost per lead climbed. You may be attracting price-shoppers with a discount-led headline. Test an outcome-led headline instead of a price-led one and compare cost per qualified lead, not just cost per click.
Ads keep getting disapproved for unverifiable claims. Google requires proof for superlatives ("best", "#1") and time-bound offers. Swap unverifiable claims for verifiable numbers you can back up on the landing page.
Tools and resources
- Google Ads Editor, for bulk headline testing across ad groups
- Search terms report (last 90 days), your single best source of real buyer language
- A shared swipe file (Google Sheet is fine) logging every formula tested and its CTR lift
- If you're managing this alongside a small team, a Google Ads agency for small businesses can run the testing cadence for you while you focus on fulfilment
What to do next
Once headline formulas are stable and CTR has lifted, the next lever is Quality Score itself, specifically landing page relevance and expected CTR by device. If you want a second opinion on where an account is actually losing money before you invest more testing hours, a comparison of the best Google Ads agencies in South Africa is a reasonable place to check how account audits are typically scoped in 2026.
FAQ
What's the best Google Ads copy formula for CTR? The number-plus-outcome formula ("4-Hour Callout", "R500 Off This Month") consistently outperforms generic benefit statements because it gives searchers a concrete reason to click rather than a vague promise.
Is CTR or conversion rate more important? Both matter, but for different reasons. CTR affects Quality Score and cost per click; conversion rate affects cost per lead. Optimise CTR first to lower costs, then optimise the landing page for conversion.
How much does a low CTR cost in Google Ads? It varies by industry and competition, but a jump from 2% to 5% CTR on the same keyword set typically lowers cost per click meaningfully through the Quality Score mechanism, sometimes by 20 to 40%.
How long should an ad copy test run before I judge it? Run it until you hit at least 1,000 impressions per variant and two full weeks, to avoid day-of-week and budget-pacing noise skewing early results.
Do emojis or symbols in ad copy help CTR? Google's policies restrict most symbols and emojis in standard text ads and responsive search ads, so this isn't a reliable lever for South African advertisers in 2026.
Should every ad group use the same headline formula? No. Match the formula to intent. Urgency-driven searches need speed and availability language; research-stage searches need proof and comparison language.
Can I copy a competitor's ad copy formula directly? You can use the same formula pattern, but the specific number has to be yours and verifiable. Copying their exact claim without matching proof risks a disapproval and doesn't build trust with your own audience.
Does ad copy testing work the same for Search and Performance Max? Not exactly. Performance Max relies more heavily on asset combinations and machine learning to select what shows, so manual headline testing has less direct control, though the same specificity principles still apply to the assets you feed it.
One last thing
The single biggest CTR lift most accounts see isn't a clever headline, it's removing a claim nobody believes. "Best in Johannesburg" gets scrolled past because every competitor says it too. A specific, checkable number always beats an unverifiable superlative, and that one swap alone often moves CTR more than three rounds of creative brainstorming.
