How A Physiotherapy Scaled Google Ads Spend by Five Times While Keeping Cost Per Conversion Flat
The client ran 8 structured test campaigns — comparing landing pages, bidding strategies, geo-radius targeting (10km vs 15km), mobile-only traffic, and keyword sets — before consolidating into a single dominant “Regenesis – General” campaign. The result: spend scaled nearly 5× while cost-per-conversion held flat at RM28, achieving a 12.40% conversion rate — more than double the healthcare benchmark. The main campaign now generates 2,550 of the account’s 2,983 total conversions with a 75.72% top-of-page impression share. The only remaining constraint is a daily budget cap causing 31.95% lost impression share.
A disciplined testing programme across eight experimental campaigns identified the exact formula for profitable paid search — and then the team scaled it without breaking what made it work.
The hardest thing to do in paid search is not building a campaign that converts. It is scaling a campaign that converts without watching the cost per conversion gradually inflate as you push into less efficient territory.
The client managed to do exactly that. Starting from a fragmented set of experimental campaigns testing different bidding strategies, landing pages, geographic radii, and keyword sets, the team eventually consolidated everything behind a single dominant campaign — and then scaled spend by nearly five times while holding cost-per-conversion at essentially the same figure.
The all-time account total: RM84,195 in spend, 2,983 conversions, at a blended RM28.22 cost per conversion. The main campaign alone — “General” — accounts for RM71,910 of that spend and 2,550 of those conversions at RM28.20 each. The numbers are almost identical, which tells you something important: the testing programme worked because it was rigorous, and the consolidation worked because it was disciplined.
The Testing Phase: Eight Experiments Before a Winner Emerged
Most Google Ads accounts are built by instinct and adjusted by crisis. Someone sets up a campaign when the business needs leads, tweaks it when it stops working, and rarely has the patience to run structured experiments before committing to a permanent approach. We took the opposite path.
Before settling on the current account structure, the team ran — or inherited — at least eight distinct experimental campaigns, each designed to test a specific variable. The experiments included a landing page A/B test comparing two page variants, a mobile-only campaign to isolate smartphone traffic, geographic radius tests at both 10km and 15km from the clinic, a bidding strategy comparison (Maximise Clicks versus Maximise Conversions versus Target CPA), and a keyword expansion test introducing new search terms.
| Campaign / Test | Spend (RM) | Conversions | Conv. Rate | CPA (RM) |
|---|---|---|---|---|
| LP A/B Test | 170 | 7 | 9.52% | 25.47 |
| Mobile Only | 301 | 12 | 13.64% | 25.08 |
| Geo Radius 10km | 451 | 21 | 16.80% | 21.47 |
| Geo Radius 15km | 3,968 | 174 | 14.72% | 22.81 |
| Bidding Strategy Test | 517 | 17 | 12.32% | 30.40 |
| New Keywords Expansion | 508 | 13 | 8.50% | 39.08 |
| Year End Promo | 2,020 | 72 | 14.72% | 28.06 |
| Target CPA Variant | 1,258 | 58 | 15.93% | 21.70 |
Several patterns emerge from this data. First, geographic targeting matters enormously: both the 10km and 15km radius campaigns outperformed the broader keyword-only approach in conversion rate, suggesting that patients searching for physiotherapy are far more likely to book when they know the clinic is nearby. The 10km radius achieved 16.80% — the highest conversion rate of any test campaign.
Second, mobile traffic converts well. The mobile-only campaign’s 13.64% rate suggests that patients searching on their phones are motivated — they are likely in discomfort and looking for immediate help, which is exactly the high-intent profile that physiotherapy advertising wants to capture.
The 10km radius campaign achieved a 16.80% conversion rate — the highest of any test campaign. Patients searching for physiotherapy near them are not browsing. They are in discomfort and ready to book. That insight shaped the consolidated campaign that followed.
Third, new keyword expansion came at a cost. The expanded keywords campaign had the lowest conversion rate (8.50%) and the highest CPA (RM39.08) of the test group — which is what you would expect when reaching beyond a proven keyword set into adjacent, less-qualified searches. Some of those new keywords may have been worth keeping. Most were probably noise.
The Consolidation: What “General” Actually Is
The main active campaign is not a single campaign in the conventional sense. It is the distilled product of everything the testing campaigns learned. The bidding strategy (Maximise Conversions) was validated through the strategy comparison test. The geographic approach incorporated radius targeting insights. The keyword set was refined by removing the terms that the expansion test showed did not convert.
The result is a campaign that has processed 392,611 impressions, generated 20,561 clicks at a 5.24% click-through rate, and delivered 2,550 conversions at RM28.20 each — with a 12.40% conversion rate. That rate is exceptional for a healthcare campaign. Most healthcare advertisers would consider anything above 5% a strong result.
Context for the 12.40% conversion rate: Healthcare paid search campaigns in competitive markets often achieve conversion rates in the 3–6% range. Our client at 12.40% is more than double that benchmark. The most likely explanation is a combination of highly specific local intent keywords, a landing page that addresses patient anxiety about physiotherapy effectively, and a clear, low-friction booking call to action.
Five Times the Spend, Same CPA: Why This Is Unusual
The relationship between spend and CPA in Google Ads is rarely linear. As budgets increase, campaigns exhaust their most efficient audiences first and then reach further into less-qualified territory. This is sometimes called the “efficiency cliff” — and most advertisers encounter it as they scale.
Comparing the testing phase (all experimental campaigns combined, approximately RM12,300 spend, 433 conversions, RM28.40 CPA) to the consolidated main campaign (RM71,910, 2,550 conversions, RM28.20 CPA) reveals something notable: the CPA did not move meaningfully as spend increased nearly sixfold.
| Phase | Spend (RM) | Conversions | CPA (RM) | Conv. Rate |
|---|---|---|---|---|
| Testing Phase (combined) | 12,300 | 433 | 28.40 | ~12–16% |
| Consolidated — General | 71,910 | 2,550 | 28.20 | 12.40% |
| Change | +484% | +489% | Flat | Stable |
The reason CPA held flat despite a near-five-fold budget increase comes down to the quality of the testing methodology. Because the consolidation was built on validated signals — not guesswork — Google’s algorithm had strong data to work with from the start. It was not learning from scratch when the budget increased; it was extending a pattern it already understood well.
This is the argument for doing testing properly before committing to scale. The investment in eight smaller experimental campaigns paid for itself many times over by eliminating the CPA inflation that typically accompanies aggressive spend increases.
The Services Related Campaign: A Cautionary Counterpoint
Not every campaign in this account succeeded. The Services Related campaign — a broader keyword approach targeting physiotherapy-adjacent queries — generated 1,636 clicks on 44,719 impressions but achieved only a 2.57% conversion rate at a RM52.95 CPA. That is roughly double the CPA of the General campaign for less than a quarter of the conversion rate.
This campaign is now paused, which is the right call. But its numbers illustrate what happens when a campaign targets intent that is too broad — when the keywords capture people who are researching physiotherapy rather than actively looking to book an appointment. High impressions, respectable clicks, poor conversions. The traffic arrives; the intent does not match what the landing page is offering.
The contrast between Services Related (2.57% CVR) and the main General campaign (12.40% CVR) is one of the most instructive comparisons in this entire case study. Same account, same bidding platform, roughly similar time period — and a five-times difference in conversion rate. The difference is entirely in how precisely the keyword intent was matched to the patient’s decision stage.
What Comes Next: The Budget Ceiling
The General campaign currently has a daily budget of RM135 and is flagged as Eligible (Limited) — specifically limited by budget. The campaign is losing 31.95% of available impressions to spend constraints.
At the current CPA of RM28.20 and conversion rate of 12.40%, recovering those lost impressions would mean approximately 32% more conversions for a 32% increase in budget. That is a straightforward, linear return — the kind of calculation that should be simple to justify to any decision-maker who can see the underlying numbers.
The story in one sentence: A physiotherapy clinic proved its paid search formula through systematic experimentation, then scaled it fivefold without degrading performance — and the remaining growth constraint is a daily budget cap that the conversion economics clearly support removing.
Physiotherapy in Malaysia sits at an interesting intersection of increasing awareness and still-developing consumer behaviour. Many patients who would benefit from physiotherapy treatment are searching — but they need to find the right provider at the right moment. An account with a 12.40% conversion rate and a sub-RM30 CPA is already positioned to capture that patient when they are ready. Extending that campaign’s reach by resolving the budget constraint is the single highest-impact next move available to the business.
