So you’ve had ad success and now your ads have stopped performing. Trust us, when we say, we understand. This can be very frustrating, whether your ads have had a slow burn downwards, an immediate drop-off, or just a rollercoaster of performance and emotions.
The good news is, there are answers to all of these situations and you can find the exact point where the issue lies and test your way out of it. Today, we are breaking down exactly how we do this for our clients and how we are able to manipulate metrics to get the results we want.
This way you can anticipate any Facebook drop-offs in performance or rollercoaster rides. Let’s get into it!
Real Rubato Client Example
Website/Account History Document
To help you see how this is done, we have pulled real data from a real client and will walk through this client’s 6-7 month process of getting their ads back on track.
The Website/Account History document helps us lay out the different time periods the client has gone through and how their ads were performing in each period.
As you can see above there were 3 distinct time periods: August-October, November, and December-Current. This highlights the past (August-October), where the client had great success with their ROAS (Return on Ad Spend), near past (November), where the client lost their account manager and had to manually take over, and the current (December-Current), where the client hired an agency to take over again.
With the new agency, they still weren’t seeing the results they wanted. Knowing what was possible, and having goals of hitting a 3 ROAS, they turned to us! Let’s take a closer look.
So what does this mean from a Media Buyer’s perspective?
There are 2 things we start with when it comes to turning around performance and that is “what story is the data telling?” and “what has changed month over month?”
To answer this, we looked at any correlational metrics on the sheet to guide us on what needs to change. Here are some things we noticed right off the bat.
- Average Order Value & ROAS: The Average Order Value of their customer’s purchases has increased which explains the sudden drop in ROAS from August to December. The more expensive the orders, the less there are of high quantity orders
- Conversion Rate & Average Order Value: As mentioned above, with the high order values, going up around $15 from period 1 to period 2, there will naturally be fewer conversions. As it shifts into period 3, the conversion drops significantly from over 7% to 1.76%, which is difficult to bounce back from
- Click-Through Rate (CTR) & Conversion Rate: One interesting thing is that the Click-Through Rate steadily increases from period 1 to 3, but the Conversion Rate actually decreases drastically. This indicates that the quality of the traffic for this product is low because even though people are clicking, they aren’t buying. We would first look at the audience strategy, account structure, and types of campaigns being run. If these change, that can create issues like these.
- Add to Cart (ATC): The Add to Cart rate also increasing from period 1 to 3 also confirms that the quality of traffic is low. They are driving more traffic, but not engaged traffic that wants to buy.
- Cost Per Milli (CPM): The last thing that catches our eye is that the Cost Per Milli is almost double from period 1 to period 2. The CPM goes from $6.1 to $13.3, which shows that the client is paying double for every impression. Even though the traffic and metrics have gone up, it is not enough to double the CPM.
Overall story: The client has traffic coming in, but it is not the right audience. Most metrics have increased but overall, this isn’t the right kind of increase, which is causing a decrease in ROAS
Funnel Optimizer Tracker Document
After taking a look at their base metrics, we start the testing! With our Funnel Optimizer Tracker, we are able to manipulate the numbers and see how exactly we can get to the client’s goal.
Note: To see the live changes, we recommend checking this out on our YouTube video and reading along!
This view takes a look at the same data, it is just filtered to the last period, period 3 (December-Current) where the client was working with an agency and not seeing the best results.
This tracker is built with formulas so that when you change a metric’s cell, the rest of the metrics update accordingly, so you can predict what your performance will look like.
To begin, and target increasing the ROAS, we tested out changing the CTR (Click-Through Rate) from 1.25% to 2% and the ROAS increased from 1.21 to 1.93. However, this may not be likely, because the client was not reaching quality users.
So let’s look at where we are versus where we could be with changes to ATC, assuming we start targeting a quality audience.
- If the ATC rate, an important metric, changes from 8% to 20%, that would take the ROAS to 2.88 which is closer to the goal. You may ask, how do you do that? There are 2 angles. We can strive to get more quality traffic, and then see what friction is on the landing page.
- If a 20% ATC is too unrealistic, we can start with 15%. This still enables the client to increase the CTR to 1.4%, producing a 2.42 ROAS, which is also closer to the goal as opposed to under 2% with an 8% ATC. The Conversion Rate would be 1.76%, which we could hopefully get up with better audiences.
- If we get a higher Conversion Rate of 4%, then the ROAS goes to 3.08 = the goal!
ROAS Calculator Document
The last step is looking at our ROAS Calculator and for each platform (i.g. Facebook and Google in this case), we break down real data for each level, the spend for each, the ROAS for each, as well as an overall ROAS.
We total this between platforms to get the final ROAS, which is 2.14. We thought we needed to get them to 2.5, their overall ROAS goal, but they were really happy with the cold and new to brand users (with an almost 70% split on cold to warm customers as well as branding, DSA, and Shopping tools).
So as you can see, all the answers are here in the data! It really is possible and we want to make you as confident as you can be with your media buying. Have any questions? Shoot one down here to ask us!