You will be able to choose a bid strategy that fits the amount of conversion data you have.
When Priya built her Google campaign, the bidding screen offered her a menu: maximise clicks, maximise conversions, target cost per action, maximise conversion value, target return on ad spend, and an option to set bids by hand. Meta's screen offered something similar under different names. Each option sounded reasonable. None came with a clear rule for when to use it.
The rule is simpler than the menu suggests. The right bid strategy depends mostly on one thing: how much reliable conversion data the platform has to work with.
In manual bidding you set the bid yourself, such as the most you will pay for a click on a keyword. Google still offers it for Search campaigns. It gives you direct control, but you are making a decision for every keyword that the platform could make for every single auction, using far more information than you have, such as the device, the time, the location and the person's recent behaviour.
Automated bidding hands that decision to the platform. You tell it what you want, and it sets a bid for each auction based on its prediction of how likely that particular person is to convert. This is the estimated action rate from lesson 1.1, The highest bid does not always win, put to work on your behalf. On Google these strategies are often called smart bidding. Meta and TikTok use automated bidding by default for most objectives.
The main automated strategies come in two kinds. Some ask the platform to get the most results it can for your budget, such as maximise conversions on Google or the highest volume option on Meta. Others give it a target, such as a target cost per action, where you set the average cost per conversion you want, or a target return on ad spend, where you set the revenue you want back for each dollar spent. Names change, so check each platform's current list.
An automated strategy is only as good as the conversion data it learns from. If tracking is missing, it has nothing to optimise towards. If tracking counts the wrong thing, such as page views labelled as leads, it optimises towards the wrong thing very efficiently. If tracking double-counts, it thinks it is doing better than it is and bids too high.
This is why module three came before this one. With a tested conversion from lesson 3.4, Install and test your tracking, automated bidding has a solid base. Without it, it is optimising on guesses.
Volume matters too. A campaign that records a handful of conversions a month gives the system little to learn from, and target-based strategies in particular may struggle. In that case, a maximise strategy without a target is usually the better start, or optimising for a more frequent event just before the sale, as lesson 2.1, The objective tells the platform what to find, suggested.
Meta and TikTok also offer controls that cap what you pay. A cost cap asks the platform to keep your average cost per result around a figure you set. A bid cap limits the bid in each individual auction.
Caps are useful when you have a hard limit, such as the break-even figure from lesson 8.1, Start from what a customer is worth. But they come with a risk. Set a cap below what the market will accept and the platform simply wins fewer auctions. Delivery slows or stops, and you may spend almost none of your budget. A campaign that spends nothing has not saved you money. It has told you nothing.
If you use a cap, start near your real break-even point rather than at a hopeful figure, and watch spend in the first days. If the campaign barely delivers, the cap is too tight.
When you do set a target, take it from lesson 8.1, not from a feeling.
Priya's maximum cost per trial booking was S$100, from a S$300 maximum per student and an example close rate of one in three. That becomes her target cost per action on Google, once her tracking has recorded enough bookings.
Farah's sale is better expressed as a return on ad spend, because order values vary. Her margin is an example 45 percent. To break even on a first order, the profit from each sale must cover the ad cost, so ad spend can be at most 45 percent of revenue. That is a return of S$1 divided by 0.45, about S$2.22 in revenue for every S$1 of ads, or 222 percent. Anything below that loses money on the first order. She might set her target a little above it, to leave a margin of safety.
Once a target is set, change it in small steps and give each change time to settle. Big jumps can send the campaign back into learning, which lesson 8.3, The learning phase: why results wobble at first, explains. And if the platform consistently cannot hit your target, it is telling you something about your offer, creative or market, not just your bid.
Decide now what you will use for your first campaign on each platform. In the activity below you will choose a bid strategy and write why.
Choose a bid strategy for your first campaign on each platform you plan to use and write why.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).