How to analyze your Facebook (or Instagram) ads campaign

After launching my Facebook course I noticed that a lot of people wanted to know how to analyze a campaign. They want to know how to judge the success/failiure of a campaign and what to do about it.

The process of running ads is pretty simple:
1. Test
2. Analyze
3. Scale

In this post I’ll talk about the second part. How to “look” at campaigns and find the best-performing ones asap. If you’ve run campaigns before, you have a baseline to compare to. If this is your first campaign, I’ll do my best to show you how to debug your campaign, or find out what might be the issue.

Shameless plug: I have just launched a new course that’s all about Facebook Ads & Facebook Marketing. I cover this campaign-analytics there in detail. If you want to get the course on Udemy, here’s a massive discount only for the readers of this blog. I have a 30-day, money-back guarantee so you can go through the entire course and get your investment back if you don’t find it valuable.. Click here to get the 94% discount on the course.

Most of my day is split fairly evenly between the three phases of Test, Analyze and Scale. Over time I’ve learned to be less judgmental over the content. Meaning that I’m willing to test almost anything. From my experience, most of the campaigns will flop, maybe 90%. The next 8% will do okay and then 2% of the campaigns really blow up (in a good way). These moments happen every few months. You find something that works so well you can double your business for example. This spike will eventually flatten out, but the plateau is significantly higher than before.

So to put things into perspective, vast majority of campaigns will under-perform and will be turn off. This is who I would analyze a campaign.

Step 1. Test campaign launch

Let’s assume that we have just published some campaigns. Each campaign has 1 ad-set and 1 ad. (I like to test 1 ad per campaign because I think it gives me a clear understanding of how that one ad performs, versus launching 1 campaign with 30 ads in it.) These campaign could be launched with a span of a few hours in one day and could be variations of creative, targeting, placements, etc.

Step 2. Watch Impressions, Cost per Result and CPM.

Impressions: In the first couple of hours after launch of the campaign, I will start seeing the first couple of impressions. The amount of impressions you get depends on your daily budget. The higher daily budget you have, the more impressions you’ll get sooner. You can spend less money, but wait longer to get the same result. Facebook recommends that a creative should have at least 8.000 impressions before the algorithm has optimized. From my experience 1.000 – 2.000 impressions is enough.

Cost per Results: Within the first 1.000 impressions I will see a few results (web conversions, app installs, etc.). The price of these few results is already an indicator of how well this campaign is doing. This means that just a few hours after launch, I already have a rough idea of how this campaign is doing.

Historical data for Cost per Result on a Facebook campaign, using website registration as the optimization event.

CPM: The cost per thousand impressions is an indicator of what it will cost me to show this campaign to people in the future. It’s a good number to look at together with the Cost per Result.

Step 3. Adjust budgets

This is the important step and the difference between how most people run campaigns. By this time I have a rough idea how the campaign is performing. It’s gotten a few thousand impression and a couple of results. If the Cost per Result is extremely high (compared to my normal baseline), I will lower the budget or turn off the campaign all together. If the Cost per Result is very low and I’m confident there are enough impressions so that it’s not just a lucky few hits, I will increase the budget a little bit. The most important thing with budget increases is to do it incrementally, don’t 10x it.

You can adjust your budget on the campaign level or the ad-set level, depending on your set-up.

Step 4. Keep adjusting

Facebook will recommend you to set a daily / lifetime budget and let the campaign do the magic. Another approach is to constantly monitor your Cost per Result on all campaigns and adjust the budget, even every hour or two. I’ve seen success with both approaches. If you’re only spending up to a few hundred dollars / day, it’s probably better to not mess with he algorithms much. Once you start spending thousands of dollars / day, I believe you can get a better return if you increase / decrease the budgets couples times a day. I was specifically advised against this much “budget manipulation”, but I think it works.

Think of this more as stock trading. The prices per result is going to fluctuate every few minutes so I think it’s appropriate to match your budget to the Cost per Result. If Cost per Result is low, keep increasing the budget. If the Cost per Result starts going up, start to decrease the budget slightly.

Shameless plug: I have just launched a new course that’s all about Facebook Ads & Facebook Marketing. I cover this campaign-analytics there in detail. If you want to get the course on Udemy, here’s a massive discount only for the readers of this blog. I have a 30-day, money-back guarantee so you can go through the entire course and get your investment back if you don’t find it valuable.. Click here to get the 94% discount on the course.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s