Learning from your social media ad results... and how Feed can help.
Updated: Aug 4, 2020
It’s tough to grow a following organically these days, so using some form of paid promotion to reach more people online is now pretty standard practice.
Luckily, social media advertising is far more accessible and affordable than many other options, such as PR, TV advertising or radio promo. Anyone can use it to promote their work… which also means anyone can throw money at social ads and achieve some sort of result. The challenge is to spend effectively, and to interpret and learn from the results of your paid promotion.
In this article, I want to focus on what to do when the results of your social media ads start coming in.
Everyone I’ve spoken with - Feed users and others - seems to face the same issues, whether you’re a total novice or a seasoned Ads Manager pro:
Data overload! There is a LOT of data out there. How can you distil the useful data?
Marketing jargon. Do you know your impressions from your reach? Do you need to, and how can you best interpret results?
And… now what? You have some results from your ads. What should you do with this information?
To summarise, in an ideal world we want to: distil, interpret and then act.
There are lots of ways to address these steps… and I’m guessing you don’t have time to read through a digital advertising opus. So I’m going to outline a basic approach that should work for people just getting started, or those who have a bit more experience.
Define your objectives
You need a benchmark against which to measure success. What do you want to achieve with your ads? Do you want to direct as many people as possible to your website to buy merch, or a product? Do you want to grow Spotify followers and listeners? Do you want to build a retargeting audience? Are you interested in audience growth across multiple social platforms?
When you start running ads, it makes sense to take a broad approach and then hone in from there. A great place to start is with a Facebook (& Instagram) lookalike audience. You don’t need to go hyper-targeted at the start (“I’m pretty sure I have a strong following in the Victoria Park area... 🤔”) especially if you’re in the earlier stages of building your audience. There are probably people in lots of areas, of varying ages and interest, who will engage with what you do.
So you start running a few broadly-targeted ads, you start getting results back... and then (understandably) many people get stuck at this point. The volume of data you get back from social media ads can be totally overwhelming.
Focus on the things that matter...
Go back to your key objectives. If you want to get people onto a website, you should be looking at click metrics. And specifically outbound clicks (ie. off the platform). The other basic metric that will be useful for any objective is cost - how much did you spend on that ad? Divide cost by number of outbound clicks and you get cost per outbound click. The next logical step would be to check results on the platform you’re pointing people towards (eg. listeners on Spotify, or purchases on an e-commerce platform). You can also do some clever stuff with Facebook pixels as well to track what people are doing on your site (we’ll cover this in more detail down the line).
Data on how people are engaging with your ads (and cost per engagement) are also relevant to understand how fundamentally interesting and enticing people are finding your content.
… to avoid analysis paralysis
Don’t get too distracted with metrics that aren’t as relevant to your objectives. It’s far better to have a few key metrics in mind - and interpret them thoughtfully - than analysis paralysis induced by a huge dashboard of 25 metrics that you can’t possibly keep track of.
I see lots of people getting hung up on demographic data. For example, “75% of my audience are aged 18–25” could be interesting data. But it could also be telling you something you already know, if your target audience is that age bracket. And if you’re not targeting that age bracket you could interpret this in two ways: “I have an untapped older audience I should be trying to reach”… or “maybe I should focus on 18–25 year olds as they seem to be responding better”. It’s often shades of grey when you’re interpreting data like these.
Focus on your content first...
From those key metrics, you will get some immediate insight into what posts people are engaging with the best. This is often different for promoted versus organic posts: what your followers like might not be what new people engage with, at least initially. What works best for each group: rough-and-ready or professional looking content, images or video, behind the scenes or finished product, talking to camera or more abstract?
From what I’ve seen, using data to refine what you’re posting about can yield much greater and more immediate improvements in ad performance than focusing on refining other things (like target audiences). I don’t know why, but this often seems to go overlooked.
For the best results you need to try out multiple posts over a longer period of time (Joshua has previously written about the benefits of a longer-term approach). We’ve talked before about often being surprised by what gets people interested and engaged - so go with a process of trial and error, and throw in some random stuff that you might not usually promote.
… and then think about other things
Looking at the results on individual posts can be informative (certainly for evaluating content) but beware of focusing too much on an immediate uplift in followers. You need to think longer term, about the marketing funnel. People don’t become your most loyal followers, listeners or customers overnight. You need to entice them first, then keep re-engaging them and developing their interest.
So look for trends over time. Has your follower trend started increasing more now you’re spending on digital ads (over the longer term, not just a few days or a couple of weeks? If you’ve been directing people to Spotify, what is the trend in followers and listeners?
You can also break things down by audience segment. Is it getting easier or harder to entice new people? Is your retargeting audience (people who have engaged with you in the past) growing, and how are people within this audience engaging? If the engagement rate is going down, you may have to think about adjusting which new people you’re reaching.
There’s a tradeoff between more specific targeting with a higher cost of your ads. The more specific targeting, the smaller the audience, and the more it will cost you to get results. That said, perhaps that audience is a “higher quality” audience more likely to become loyal fans or customers. Drilling down into how people are behaving once they’ve clicked your ad (listeners on Spotify, or purchases on your website) can help shed some light on the cost vs quality argument. Again Facebook pixels can be useful here.
Our approach at Feed
We started building Feed with these principles in mind - understanding that people want access to great social media advertising without having to devote a huge amount of their time to marketing.
For the beta version, we wanted Feed to:
Give users straightforward ad results, easy to interpret at a glance… but allowing them to delve into the details, if they want to.
Automatically set up the marketing funnel by creating audiences (including a retargeting audience).
Give feedback on both organic performance from paid performance.
Allow users to track audience growth not just on Facebook and Instagram, but across other social media and streaming platforms.
The feedback from Feed users so far has been great. Having access even to straightforward (but easy to interpret) data changes the way people post to better suit their audience, and empowers them to make better decisions when it comes to digital marketing.
Taking care of basic principles like the marketing funnel and giving users the results that matter has proved valuable: on average, Feed generates 4x the results (in terms of engagements, clicks etc.) for a given budget than achieved by Feed users previously.