This month, Facebook is rolling out one of the biggest Ads Manager changes in its history: campaign budget optimisation.

According to Facebook, “campaign budget optimisation is a way of optimising the distribution of a campaign budget across your campaign's ad sets.”

As the platform stands currently, advertisers can allocate budgets on ad set level, meaning you can assign higher spend to high-value audiences and vice versa. This gives advertisers direct control over how much is spent, serving adverts to each audience, regardless of performance.

But Facebook is confident that its campaign budget optimisation algorithm can do a better job. Due to the immediate nature of the bidding system, the algorithm automatically and continuously scopes out market opportunities that you may not even be aware of, allowing you to make best use of your money (as well as saving time).  

This means that you’ll only be able to allocate budget at campaign level, leaving the rest to the algorithm.

Working with our partner The Next Ad, we’ve been trialling campaign budget optimisation for a while – albeit with more advanced rules than Facebook advertisers in Ads Manager currently have access to. We found the changes saved us time, but there can be some initial issues.

So here are our campaign budget optimisation recommendations, based on our experience so far:

Fewer ad sets, broader targeting and automatic placements. According to Facebook, by maximising potential reach and clearly defining your goals, it can scope out the best opportunities. This setup gives the algorithm a huge pool of target users to learn from, helping it decide how to most effectively spend your budget based on the campaign goal.

You’ll be able to bid different amounts for different audiences, so you can prioritise those that matter most to you. For example, you could bid higher for an audience you know will generate high-value conversions, and bid more tentatively for an untested audience.

Facebook will be introducing minimum and maximum spend caps at ad set level, but it has stipulated that excessive use of these controls is not recommended. However, we believe that this will be an important way for advertisers to gain back some control over ad spend.

The higher the quality of data you provide Facebook when building your campaigns, the more intelligent its decisions will be. For example, with a menswear ecommerce campaign, you could use carefully segmented CRM data to make sure you’re targeting people who have previously shown an interest in menswear, or make a lookalike audience of your top menswear customers.