Every time someone scrolls down their news feed to an ad space, or swipes through their stories to an ad, an auction is held. All the advertisers whose targeting criteria matches that user enter the auction, where they bid for the right to show their ad to that user.
The bids that advertisers submit naturally depend on their budgets, their aggressiveness, and how valuable that user is to their brand. But there's something else which plays an important role in how advertisers bid, and that's their bidding strategy.
The term bidding strategy refers to a method of controlling how much you're willing to pay to show your ad to a user. There are four main bidding strategies within Facebook Ads, all of which can be activated at the campaign level.
This article walks through each of the different strategies, how they work, and when you should be using them.
How does it work?
Lowest cost bidding allows you to specify a particular budget, and then Facebook will optimise spending within that campaign to deliver results at the lowest cost per result. Given that Facebook will try to always spend your daily budget with this bidding strategy, this is effectively the same as Facebook trying to maximise the number of results that you can get within this campaign's budget.
Facebook will try to maximise results within this campaign by shifting spend across the placements and users that it thinks are most likely to deliver results in the future. If you're using campaign budget optimisation (CBO) then it'll also allocate spend across ad sets in a way which will maximise overall results.
I'd recommend using CBO with lowest cost campaigns; if you're willing to let Facebook optimise across placements and users, then you should also be willing to let it control your ad set spends.
When to use it?
There are two main scenarios in which to use lowest cost.
The first is when you're not constrained by hitting a certain cost per result, and you just need to spend your budget. If you know exactly how much you want to spend, or you're working for a client with fixed budgets, then lowest cost is the way to go. It'll ensure you stay exactly on track to hit your budgets, while delivering the most results possible within those budgets.
The other scenario in which to use lowest cost is if you're launching a campaign. You might ultimately want to use one of the bidding strategies described further down the page, like target cost, but these require data to work with. When you're launching a campaign, you don't have that data.
A solution to this problem is to start your campaigns on lowest cost until they reach a certain number of results. Once they've passed this point, and have enough data for the other bidding strategies to work with, you can switch them over to one of those strategies.
Starting with lowest cost works because it forces a certain amount of spend through your campaign. If you start off on something like target cost, without any historical data, you'll find that your campaign is very slow to get going, and may never spend as much as it otherwise could.
How does it work?
Target cost bidding seeks to bring in as many results as possible, while keeping your average cost per result in line with a number that you specify. If you want your average cost per result to be $50, set it up on target cost and you should find that over a medium-range time period (30 days) your cost per result will average out close to $50.
Facebook achieves this by estimating each user's probability of conversion once they've seen your ad. It then submits a bid in line with this probability, and your target cost. A very simple way of thinking about this is if you have a target cost per result of $50, and Facebook thinks that a user has a 1 in 10,000 chance of converting, it'll submit a bid for that user of $50/10,000 = 0.5¢. If Facebook finds itself below your target cost per result, it'll bid more aggressively, and vice versa.
When to use it?
The most obvious case of when to use target cost bidding is when you're being judged on your ability to maximise volume at a certain cost per result. This could be because you're working for a client with a certain gross margin target, and so in order to hit that gross margin you need to deliver results at a fixed cost per result.
Target cost is great for situations such as the above, where you need to be able to predict with some accuracy your cost per result. Constraining Facebook by giving it a target cost can have implications for volume though. If your target cost is too low, you may well see low delivery. It's not uncommon to see ad sets spend less than $1 a day on target cost, when they're unable to hit the cost per result they've been tasked with getting.
One case in which you shouldn't be using target cost is if your campaigns are budget capped. If they're budget capped, there's no point telling their ad sets to bring in results for a certain cost per result. Instead, simply switch to lowest cost bidding, and let Facebook optimise your campaign to deliver as many results as possible. In most cases this will bring in more results, and bring down your cost per result.
Following on from the above, if you're using target cost you should be willing to set your budgets high. You should set a target cost where you know you'll be making a decent net profit. As long as you know that you can confidently service as many results as these campaigns bring in, then this is a very low-risk way to generate results.
How does it work?
Cost cap bidding works in a very similar manner to target cost bidding. Instead of setting a target cost that you want to aim for though, you can set the maximum cost per result that you're willing to tolerate.
Cost cap bidding will then work by exploiting all the 'cheap' results it can find within your budget. Once these are exhausted, it will gradually bid more and more aggressively, and its cost per result will rise. It will keep on rising up to the point it hits your target cost, at which point it won't raise its bid any further. From this point on, your cost cap campaign will behave similarly to a target cost campaign.
When to use it?
If you don't need to produce predictable cost per results, but know the maximum cost per result that you'd be willing to tolerate, use cost cap bidding.
Cost cap bidding combines the best of lowest cost bidding, and target cost bidding, and should be the go-to for advertisers looking to maximise profit. Simply set your cost cap to be somewhere below your gross margin per result, and let it bring in as many results as possible. It'll start off by bringing in as many cheap results as possible, before settling on your specified cost cap.
How does it work?
Bid cap campaigns work a little different to other bidding strategies, in that they're the only ones which effectively use manual bids. Bid cap campaigns require you to input a maximum amount that you're willing to pay for each result. Facebook will then translate this into an impression (/CPM) bid, based on your conversion rates, and enter auctions with this bid.
Facebook uses a 2nd price auction model, which means for each auction you win, you pay the minimum amount it would have taken to beat the 2nd price bidder. This is important because it means that the amount you pay is less than the amount you bid, and subsequently your cost per results will end up lower than the bid cap you input.
When to use it?
Bid cap campaigns offer the most manual level of control. This can be useful if you want to reach as many users as possible at a specific bid, or want to increase competitiveness on a certain audience.
There are a number of drawbacks though which mean you should stear clear of bid cap campaigns where possible though. The first is that they require a lot manual effort to optimise your bids in line with performance. This is work that can be handed over to Facebook by using another similar bidding strategy, cost cap bidding for example.
It's also difficult to predict the sort of cost per result that you're likely to see when launching a bid cap campaign. All you know is that it's likely to be below your bid cap, provided that Facebook hasn't over-estimated your conversion rates, and hence miscalculated your CPM bids from your bid cap.
Personally I'm a fan of cost cap bidding. It's an effective way of maintaining predictable cost per results, without overpaying for cheap results. If you need to throw money at an ad set though, perhaps to promote a limited offer, then lowest cost bidding is typically the way to go. I'd avoid steering clear of bid cap campaigns though, at least unless you have a good reason to be using them.