About bid strategies: Lowest cost

30/11/2019
Note: The lowest cost bid strategy used to be known as automatic bidding. The strategy with a bid cap used to be known as a manual bid set as the "maximum". Each choice functions exactly as it did before. We describe them differently now to more accurately reflect how the strategies works, and their benefits and trade-offs. We're now also testing a new "cost cap" functionality. You can learn more about this in the article, but please note that it may not be available to you at the moment.

Getting the lowest cost per optimisation event is one of the bid strategies available for Facebook ads. In addition to telling our system to get you the lowest cost per optimisation event possible, you can also set a:

  • Bid cap: This tells us the maximum amount that our system can bid for you for any given optimisation event.
  • Cost cap: This tells our system to bid for you with the goal of keeping your average cost per optimisation event at or below this amount. Please note that we're still testing this feature, so it may not be available to you at the moment.
This bid strategy is best for spending your budget as efficiently as possible.

The trade-offs of that efficiency are that your cost per optimisation event may rise over time, as your spend increases and/or as auction competition increases.

If you value cost stability more than maximising the efficiency of your budget, you can instead choose the target cost bid strategy. (Please note that that strategy is only available for lead generation, app install, conversion and product catalogue sales campaigns.)

Should I use a cap?

If you want to control your cost per optimisation event in some way, we recommend at least testing a cap – especially if some/all of these apply to you:

  • You know how much an instance of your optimisation event is worth to you
  • You know what cost per optimisation event you need to achieve to make a profit
  • You want control over the maximum cost you pay for a given optimisation event (bid cap) or your average cost per optimisation event (cost cap)
  • You have the time to monitor your results and make adjustments to your cap as necessary

How do I decide between a bid cap and a cost cap?

Note: Cost caps are being tested and may not be available to you at the moment.

It depends on your goals and flexibility.

Cost caps are more flexible, which means that they're less likely to constrain delivery than bid caps. This is because cost caps apply to your average cost per optimisation event, so our system can pursue optimisation events at a wider range of costs. While some optimisation events may cost more than your cap, over the lifetime of your ad set your average cost should be at or below your cost cap amount.

Bid caps are less flexible, which means that they're more likely to constrain delivery than cost caps. This is because bid caps limit what we can bid in every auction. If you set a bid cap of £5 and 100 optimisation events become available for £5.01 bids, we won't try to get any of them (we would be likely to go after all of them if your cost cap was £5). We recommend a bid cap over a cost cap only if you need to tightly control the cost of every single optimisation event you get.

What should I set as my cap?

When trying to figure out what initial cap to set, it may be worth considering the following:

  • Past performance data. The average cost per optimisation event that your ad set was getting when you weren't using a cap can be a useful starting point. However, if you're using this information to set a bid cap, bear in mind that bids are often higher than costs. This means that setting your bid cap at your average cost per optimisation event could lead to winning fewer auctions.
  • How much an optimisation event is worth to you. If you know how much a result is worth to you (e.g. the maximum amount that you can afford to pay for an optimisation event while making a profit from it), try starting with that.
  • Your budget to cap ratio. We recommend setting your daily budget at least five times higher than your cap so that you're well positioned to get around 50 optimisation events a week (the amount needed for our system to achieve relatively stable delivery). If you need a minimum to start with, you could divide your daily budget by five and set that as your cap. Then see how it performs and adjust as needed.
Succeeding with a cap sometimes requires trial and error. It's a good idea to monitor your performance and make adjustments if you're not meeting your cost goals. For example, if the auction gets more competitive, you may need to raise your cap to keep getting results. You may also need to lower it if your costs get too high. A willingness to test and adjust will help you find the right balance between staying competitive and controlling costs. If you can't monitor and adjust your cap, we don't recommend setting one.

Tip: A good time to check to see if an adjustment is necessary is after your ad set has got about 50 optimisation events. At that point, its delivery should be relatively stable, so you can get a pretty accurate sense of its performance. If you decide that a significant cap adjustment is necessary, make sure that you let your ad set get about another 50 optimisation events after making the change. If you don't, you're not giving our system a chance to stabilise delivery based on your new cap.

Bid cap guidance

The most important thing to bear in mind when setting a bid cap is that, although a bid and the cost of an optimisation event are related, they aren't the same thing. In fact, you may pay less than your bid for any given result as you only get charged the minimum amount that we would've needed to bid to win the auction.

It may be useful to think of a bid cap as the absolute maximum you'd be willing to pay for a result (as it's possible to be charged your full bid if the ad auction was very competitive), while bearing in mind that it's likely you'll be charged less than that for many of the results you get.

What happens if I don't set a cap?

Not setting a cap tells us that we can bid as much as we need to try to fully spend your budget. It means that certain actions such as increasing your budget significantly, or other factors such as a sudden increase in auction competition, could cause costs to rise sharply. It also means that your ad set will be optimised to spend its entire budget by the end of the day or its lifetime (depending on its budget type), regardless of how expensive a given optimisation event is. We'll still try to get you the lowest cost results available, but if the only way to ensure that your entire budget is spent on time is to go after results that are more expensive, we'll do that.

Are there any risks to setting a cap?

There can be. If you set a cap that isn't competitive, you may not be able to win enough auctions for your ad set to deliver consistently (or at all). This can make it harder to spend your full budget. If that happens, you can try raising your cap or removing it. If this doesn't work, you can also try adjusting your targeting and creative. Doing so could help you reach new people or recapture the attention of your audience, allowing you to win more auctions without adjusting your cap.

Additional information

Do cap changes affect ad performance?

The effect that a cap change might have on your results depends on whether or not your ad set is on track to spend its full budget. Here's a basic explanation of what raising or lowering a cap might do, divided into ad sets that are spending their full budgets and those that aren't.

If your ad set is on track to spend its full budget:

  • Raising its bid cap. This isn't likely to do anything unless you also raise your budget. Our pacing system will probably just lower the bid back to what it was at before.
  • Lowering its bid cap. This could lead to less delivery if you lower your bid cap to an amount that's less than what our pacing system had it at before, as you may not be able to win as many auctions with a lower bid.
  • Raising its cost cap. This isn't likely to do anything unless you also raise your budget. We're likely to continue bidding for you in the same way.
  • Lowering its cost cap. This could lead to less delivery if you lower your cost cap to an amount that's less than what your average cost per result was at, as you may not be able to win as many auctions with reduced bidding flexibility.

If an ad set isn't on track to spend its full budget (also known as "under-delivering"):

  • Raising its bid or cost cap. This could increase delivery, and we recommend trying it if you can.
  • Lowering its bid or cost cap. This could lead to even lower delivery.

Notes:

  • We recommend not making significant edits to your cap until your ad set has got about 50 of its optimisation event.
  • Since bid caps are more restrictive than cost caps, these effects are more likely to occur (and to be more severe) for bid cap changes.
  • If your ad set isn't on track to spend its full budget, our pacing system isn't lowering its bid to try to get you more results. It's entering your bid cap amount into every auction.

What's the difference between using a cost cap and using the target cost bid strategy?

Both bid strategies use an average cost as a benchmark, but aside from that, they work differently. A target cost is an amount that we try to stay close to even if there are lower-cost optimisation events available. It uses a different type of pacing than the lowest cost bid strategy to allow you to increase your budget while maintaining your target cost. A cost cap is an average amount that we try to stay under while still getting you the lowest-cost optimisation events first. We only go after more expensive optimisation events that are closer to your cost cap when we've exhausted all of the cheaper ones. If your average cost per optimisation event has reached your cost cap, it's unlikely that you'll be able to significantly increase and fully spend your budget without also raising your cost cap.

Here's a simplified example to further illustrate the difference:

Let's say there are two ad sets. They're identical apart from one using a cost cap of £10 and the other using a target cost of £10. There's a set of optimisation events available for varying costs. You could spend your full budget with an average cost per optimisation event as low as £7. The cost cap ad set will do that. The target cost ad set will actively pursue more expensive optimisation events in order to try to get as close as possible to an average cost per optimisation event of £10. This makes it easier to significantly increase your budget while maintaining that target cost. If you significantly increased the budget of the cost cap ad set, the average cost per optimisation would be likely to increase to your cap as you exhaust the lower-cost optimisation events.

Is it possible to be charged more than my bid cap for a result?

In some instances, yes. It could happen if your ad set is optimised for conversions or app installs/events (in other words, rarer results) and we overestimate how many people will get you that result before we've received enough data on its performance to learn from (about 50 conversions/installs/events). In this period during the lifetime of such an ad set, we may try to make up for the lack of data with a higher bid, which in rare instances could lead to you paying more than your bid for a result. This is more likely (but still unlikely) to happen if your ad set has a smaller budget.

* Nguồn: Facebook