Outside of adjusting your campaign budget settings, daily spend and max bid, there are also other ways to improve the ad delivery.
One of the most important ones is to avoid de-duping (competing campaigns) where you target the same audience with different ads. Updating your ad creatives periodically and refining your target audience by adjusting demographic settings also works well. Some campaign types are by their nature suitable as ongoing campaigns, such as cart re-targeting, whereas a customer acquisition campaign might run out of suitable audience and suffer of ad fatigue if the base audience and max reach is small. In such event, changing the target audience or changing the ad creative are good tools to tackle ad delivery issues.
Ad fatigue is a phenomena which occurs eventually to all campaigns, especially for prospecting. Let’s say that you run a campaign with 100€ or USD daily budget and set the time period of 21 days and use a suitable interest for the campaign, targeting your key customers based on location, gender and age.
Following the example, let’s again assume that your audience size is just 25.000 Facebook users. Once Facebook launches the campaign, it’ll try to bid for the most valuable and suitable users, who will either ignore or engage with your ads. It’s very likely that you have reached out to half of your audience in matter of a week or less, meaning that at least most valuable audience has been already exposed to your ads once or twice. Once this peak is reached, your campaign might experience fatigue as Facebook doesn’t have any good users to expose your ads to, meaning that the campaign performance starts to decline as Facebook now reaches those users who are less likely to buy. Only way to fix this is to adjust the ad’s targeting options, creative text or possibly end the campaign and launch a new one with different targeting options.
Campaign Conversion Issue
In case your campaigns are delivering and driving traffic to your site, but aren’t yielding any conversions, adjust expectations accordingly especially with prospecting. Facebook ads by their nature target possibly interested users with relatively high likelihood to shop online. Unlike Google’s search term based ads targeting users interested in something specific [based on what they search], Facebook targets users further away from the end of conversion funnel. Often average rates can be way lower than site’s average conversion, but as Nosto’s research shows, conversion-rate is not a good metric when assessing performance of ads. Should you want to make a tough distinction, Google’s search term based ads are great when you want to target customers who have made a buying decision, but look for a suitable shop, whereas Facebook is good at targeting customers based on indication of future interests and needs.
If your ads are delivering buying customers with a reasonable return on ad spend, on average 3-5X, consider increasing your budget for such a campaign regardless of possibly low conversion-rate.
Often poor performance is caused by high expectations compared to low campaign budget. For instance if you are able to buy in traffic with low cost per visit price such as 0,2€, but using just 10€/day, campaign will drive 50 visitors a day. Consequently if conversion-% is 0,5%, every fourth day someone through the campaigns will buy something. This’ll put your customer acquisition cost or cost per acquisition (CPA) at quite acceptable 40€ and assuming that average order value is 120€, ROAS would be 3X.
When searching for best performing parameters for a campaign and when testing Facebook as an ad platform, having higher budget enables you to test faster. Using the previous numbers and example, if daily spend is low at 10€ and CPA is 40€, it takes relatively long time, four days, to get a single result and even longer period that you can take as a trend of the performance. To reliably test performance, you would need to have at least tens of purchase conversions and using example’s low budget would require months to reach a level where conclusions can be made.
As a rule of thumb, think of how many orders you want in order to get an indicative estimate of the performance and thus calculate how much money you would need, and then decide when you want to look at the results. However, don’t use CPA as a guideline for your bid as this should always be value of the conversion, not average cost.
Too Small Audience Sizes: Re-targeting
In case your site doesn’t receive much traffic and you are running any of the re-targeting campaigns, Nosto and Facebook simply might not have a reasonable audience to target, causing that your ads aren’t delivering! In practical terms, if your site receives 10 purchases a day and you are running a post-purchase campaign targeting existing customers, Nosto and Facebook have only 10 new candidates daily to target who also need to be active and identified by Facebook. On average, Facebook claims that they can identify approximately 60-70% of online shoppers, but using the example figures, audience size increases very slowly just 6-7 a day.
Same goes for re-targeting cart abandoners and customers who visited your site. While these audiences are by nature bigger than those who have purchased as there are always more cart abandoners and more visitors than cart abandoners, the audience can still be very limited by its size. To summarize, often the problem with ad delivery and re-targeting is in the logic following the audience size. This doesn’t mean you shouldn’t run such campaigns, but it should command expected commercial impact.