If there’s one thing you can count on as a media buyer, especially on Facebook, it’s volatility.
There are so many moving parts that it’s impossible to predict performance. Seasonal fluctuations, algorithm updates, relevant social trends, financial trends (did someone say stimulus?), quality of current assets, the list goes on and on. All impact the day to day performance of ad campaigns and unfortunately, no one can predict these things.
It’s like someone telling you they can predict the stock market… It just doesn’t happen.
Facebook has been getting better at providing us with more and more insights to analyze what’s going on in the auction competition, audience saturation, etc. and it’s crucial to look at the data to make logical decisions to get things back on track.
From my experience, once performance begins to dip, it tends to decrease rather quickly and usually it won’t fix itself. That being said, keep in mind that volatility is normal, and it can last for multiple days due to competition and other factors that are outside of your control.
Where do you start when things start to tank?
9 out of 10 times it’s an issue with creatives and things need a good ol’ fashion refresh.
There’s a few reason’s your creatives might be getting exhausted:
1. You’ve reached a significant portion of your audience with the same creatives and they lose their effectiveness.
You’ll know this is a culprit if you notice your frequency starting to climb.
2. You’ve reached a large portion of the “top tier” of your audience with your creatives and the algorithm starts showing it to less relevant people.
Facebook prioritizes people differently within the same audiences based on relevance, and likelihood of your desired outcome (conversion, click, brand recall lift etc.)
Among other things like bid and ad quality, the relevance plays a massive part of who sees what ad. The more relevant your ad is for a user, the better the chance of them seeing your ad first. There are a few exceptions to this but it’s a good way to simplify it.
This is why oftentimes campaigns will perform very well in the early days after launching and on a low budget. If it’s relevant to a few people, the algorithm will find the low hanging fruit and performance can be really good.
But when you go to scale it up to a broader audience, it’s not always effective and the algorithm looks at KPIs and determines that your ads might now be as relevant as someone elses and in a sense you get penalized.
The better your ads, the longer they will perform as they will resonate with more people in your audience. But once that relevant audience sees your ad, the effectiveness of your message with significantly decline. And since social media platforms are content driven (particularly visual), refreshing the creative is the most common solution to improve performance issues.
It’s fairly easy to determine if this is the issue, because you can isolate KPIs that are directly correlated to the creative/on-Facebook metrics, like your click-through rate, average watch time, attention rate, frequency, etc.
Before you jump to conclusions, take some time to troubleshoot.
Before assuming the creatives are exhausted, we first take a few minutes to see if there’s a bigger underlying issue like poor quality traffic (it happens), increased CPMs (but stable CTR), or other signs that there are outside factors affecting performance.
Here’s a list of things to check:
Website conversion rate – check to see if the C.R. is fairly consistent over the period you notice a drop in performance (look at a macro level, not day to day).
If you’re running other traffic channels, check to see if the performance on all channels is dropping, or if it’s a platform specific issue.
Check you CPM, and auction competition insights, to see if its competition causing the issues.
Look at your click-through rates and % change in those metrics.
Look at your average watch time and % change in those metrics.
Look at the average attention rate and % change in those metrics.
From the above, you can get a good indication of where things are falling off, and whether its a creative refresh that’s needed, or something that’s broken at a different stage in the funnel.
There’s an old saying that “you can’t fix what you can’t measure”.
Being able to efficiently troubleshoot and solve the problem is a common skill among top media buyers and it takes time to get the knack for it, but it helps to have a clear dashboard to read data and see performance on a day over day, week over week, and month over month basis.
It can be tough to do this from within the ads manager, we prefer to aggregate data in Google Sheets, using Super Metrics. It makes it easier to read data and we can analyze KPIs in comparison to other traffic sources as well, like Snapchat, Google etc.
If all channels are decreasing in performance simultaneously, chances are good there’s a core cause outside of say, your Facebook ad creatives.
If you determine that creative fatigue is causing the decrease in performance, here are some steps you can take to minimize/counteract the volatility.
1. Test your best creatives with new audiences. – Introducing creatives that have proven to be successful to a fresh set of eyes can be extremely effective. Just keep in mind that eventually ALL creatives will lose their scaling power so this is not a long term solution.
2. Repurpose past content to give it a fresh new look/feel. – Even small tweaks to an existing ad can extend the lifetime of your creative assets. Changing the leading clip to improve the attention rate is going to have the biggest impact so consider starting there.
3. Source entirely fresh content. – Even the best ads in the world with multi-million dollar budgets lose their spark. A message can only be seen so many times before it becomes old news, and there’s a reason the big brands launch new TV commercials every 1-2 months.
It really is essential to have people constantly watching trends, studying consumer behaviour and executing on content that’s likely to resonate with your audience.
4. Create a warchest of content and post 2-3 times per day per platform and analyze organic performance. You’ll get a lot of insights into what your audience likes best.
As an eCommerce brand, having a great product is only part of your business. You’re also in the content creation business and having content to fuel your growth is essential to hitting your scale targets and minimizing volatility.
Where a lot of people go wrong is investing all their resources in a few creative assets. Whether or not those assets perform well, you’re still throttling growth because that content has a lifespan.
People often say quality is better than quantity, however I disagree. Both are equally important, especially on the paid media side of things. You need to follow the data and let the market tell you what to push.
In the same breath that we always want to be testing new content, you should also make an effort to test other aspects of the customer buying journey as well as retention funnel. Content is just one piece of a larger picture, albeit, and extremely crucial piece.
Marketing is a never ending cycle of optimization, trial, and error. Volatility is part of the game but making sure to be proactive in trying new things even when performance is at it’s best, is what allows some brands to see exponential growth and maximize consistency.
What if it’s not a creative issue?
Not all volatility is due to creatives. Competition fluctuates with seasonality, incognito bidding wars, and other factors as well. There are times when the click through rate on creates is amazing but the website conversion rate is absolute dog shite.
Sometimes, the quality of traffic from certain platforms is just plain bad.
While the quality of your advertising DOES play into the quality of traffic, it’s not always the only factor. If you feel like your creatives are performing very well, and the only difference is the traffic just won’t convert on a proven landing page/funnel, we recommend refreshing your audiences.
We will just take our top performing lookalike audiences and recreate them on slightly different time frames as well as test new interests as well. We do recommend keeping audience sizes as big as possible, (2 million +), unless you enjoy spending all your time in the ads manager.
We’ve had a lot of success lately with broad audiences and not really using any targeting at all. Keep in mind that the algorithm will still prioritize “pockets” of traffic from within the massive broad audiences, so if you duplicate 3 broad audiences, chances are good some will perform much better than others.
Same goes for big interest/lookalike audiences so make sure you’re not shying away from testing multiple of the same audience. When you go to scale, however, running big budgets to the same audience will cause overlap, and performance will drop on certain audiences. The idea is to test a ton of stuff (creatives, audiences, copy, landing pages, etc) on a low budget and scale based on KPIs.
Hopefully you got some inspiration from this article if you’re experiencing performance fluctuations. One thing I want to emphasize is that there are thousands of ways to structure ad campaigns and there’s not really a “right” way to do it.
All that matters is that you’re taking action and running things with a purpose. Think, hypothesize, test, analyze, repeat.
Take care y’all!
Ryan
