You will be able to increase spend on a winning campaign while keeping results readable.
Priya's Google Search test worked. Over four weeks it brought trial bookings at about S$75 each, comfortably under her S$100 maximum, and about a third of the families enrolled, to use example figures again. The obvious next move was to spend more. So she raised the daily budget from S$20 to S$100 one Monday morning, and by Friday each booking was costing well over S$100 and the campaign was showing a learning status again.
Nothing was wrong with the campaign. The way she scaled it was the problem. Scaling is where many good campaigns get broken, and it does not have to be.
A large, sudden budget change is one of the edits that can send a campaign or ad set back into learning, as lesson 8.3, The learning phase: why results wobble at first, explained. The system suddenly has far more money to spend than it has learned how to spend well. It widens delivery quickly into people and auctions it knows less about, and costs often jump while it adjusts.
Raising the budget in steps avoids most of that. Increase it by a moderate amount, wait several days for results to settle, check that cost per result is still within target, then raise it again. Each platform's help pages describe how it treats budget changes and what it counts as a large one, so check the current guidance rather than relying on a rule of thumb from someone else's account.
Priya's second attempt went up in steps over four weeks, with a check before each one. It took longer to reach her new budget, but the campaign never lost what it had learned.
Even strong ads tire. As the same people see the same ad again and again, fewer of them respond, and the platform's predictions of response fall with them, a pattern advertisers call creative fatigue. You see it as a slowly rising cost per result, a falling click rate and, on Meta and TikTok, a rising number of times each person has seen the ad.
Scaling speeds this up, because a bigger budget shows the ad to the available audience more often. So when you scale, plan new creative alongside the winner rather than waiting for it to die. Add new ads to the existing ad set or group, built on what made the winner work, such as the same angle with a different hook or the same offer shown by a different customer. Let them compete with the winner instead of replacing it.
Adding new ads can itself count as an edit that affects learning on some platforms, so add them in planned batches rather than drip-feeding one a day. On TikTok, where the feed moves fastest, fresh creative tends to be needed more often than on other platforms.
Budget is not the only way to grow. You can also expand into new keywords, new audiences or a new platform. Each brings new people, and with them uncertainty.
Expand in one direction at a time, for the same reason you test one thing at a time in lesson 9.1, Test one thing at a time and decide in advance. If Priya adds new keywords, raises the budget and launches on Meta in the same week, and costs rise, she cannot tell which change caused it.
A sensible order is often to grow what is already working first, then add closely related keywords or audiences, then try a new platform with its own test budget and plan. Treat each new platform as a fresh test, starting again from lesson 6.5, Build your first social campaign, or the Search plan in lesson 7.5, Plan your Search campaign.
Even with careful scaling, expect cost per result to rise as you spend more. The platform shows your ads first to the people it predicts are most likely to act. As the budget grows, it has to reach further, to people who are a little less likely to act. Each extra dollar buys a little less than the one before.
That is normal, and it is why a campaign that looks brilliant at S$20 a day may only be good at S$100 a day. The question is not whether costs rise but whether they stay below your maximum from lesson 8.1, Start from what a customer is worth. Watch cost per result after every step. When it approaches your maximum, you have probably found the size this campaign can profitably reach with its current offer and creative. To grow beyond that point, the offer, the creative or the landing page has to get better, because more budget alone will only push costs past your limit.
Priya's campaign settled at a higher budget with bookings at around S$90 each, still under her limit, and she held it there while she worked on a second landing page.
Pick a campaign that is hitting its target, real or imagined. In the activity below you will write a four-week scaling plan with budget steps and dates for new creative.
Write a four-week scaling plan for a campaign that hits its target, with budget steps and new creative dates.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).