So many metrics, what should I measure on social media?
Chances are, if you ask 10 people, you’re going to get 10 different answers. The best one is — it depends…
First of all, measure everything. Strategies, goals and tactics change over time. Measure everything now, and worry about what you’re going to do with the data at a later date — you don’t want to start fresh every time.
Brand new company, no audience
Learn the basics. At this point, you likely have minimal content and minimal data. Data is incredibly valuable and you can do a lot with it, but at this stage there really isn’t any point in spending too much time analysing every little thing (there are other things that you should be spending your time on and the data set is likely too small).
- Create content — unless you have at least 30–45 pieces of content, you’re not going to be able to do anything. You’re a brand new company and you need to establish to the world that you’re legitimate and not going to run away with someone’s money. This is the time to share how you’re different and why people should pay attention to you.
- Build an audience — you need an audience of at least 1,000 before you’re going to be able to move on to the next step. My suggestion here is to not include your friends and family at this stage. Yes, they’re incredibly supportive and it’s better to have some engagement than no engagement, but unless they’re your target audience the ‘likes’ here are a little meaningless (they’re being nice by following your brand, but they won’t interact like real fans would). In the grand scheme of things their follow-but-no-engagement could be hurting your brand (say you have 10 fans but none of them interact? Doesn’t look good in the eyes of the Instagram algorithm).
While the importance of data is small here, it’s still important to track as it gives an indication of what’s working (note this is an indication and not necessarily causation).
You should worry about two things at this stage — audience size is one — and it should grow (even if slowly) and at this stage, audience growth will most likely take place offline (e.g. events, gatherings, community meet ups). The other thing is content volume — make sure you are able to create content on a regular basis (twice a week for example). If you’re unable to create content regularly, social media should be tackled when you have time and energy to do so at a later stage (a page that updates monthly is better than one that updates sporadically every few weeks).
A little traction, what next?
So you have a bit of a fan base (1,000+ fans) and post regularly (for example, twice weekly). At this point you should be creating content that does well, does poorly or just about average. You should also have a decent number of comments every time you post (if not, where did you get these fans?). Now you get to test out things and play around a little.
At this point you’re going to want to have a plan in place and measure new metrics (potentially tagging the type of content you’re posting — colours, text of graphics, borders, mood, etc) — and keep doing what works.
- Experiment with content — Within content you can do a lot. This is the stage where you take a longer approach and try to understand what content can do for you — is it trying to convey a message/overturn a misconception? Is it trying to expand your brand/repertoire? Is it trying to push your audience to do something? Tracking your content allows you to make judgements on what direction your social media should be taking (too often this approach is haphazard and theoretical) — set out with theories, measure and test them out.
- Experiment with timing/frequency — there is material that you can google on when and how frequently you should post on Instagram/Facebook. These are very general and they don’t necessarily have your audience in mind (why would a brand selling TVs post in the same manner as a media company? Makes no sense). By looking at your audience demographic data (readily available on the native social media channel apps) you should be able to make educated guesses on how frequently you should be posting.
- Metrics to look out/watch out for — track one thing at a time. If you’re only worried about growth, or only worried about engagement, then track the right metrics. Very often there are secondary effects to your actions so bare these in mind as well (trying to improve engagement will have a knock-on effect on overall social following). There is nothing wrong with tracking basic metrics, what’s important is that you understand why they’re moving and changing (and not just because you’re blaming the Facebook algorithm changes).
IF you only have capacity to track one thing, make it the hardest thing for your fans to do— for Facebook this means tracking ‘shares’ (which is harder than ‘comments’, which is harder than ‘likes’). Thinking about how to increase your ‘shares’ on Facebook will mean only making great content that can work harder for you — it takes a lot of willpower to ‘share’ something publicly with all your friends whereas ‘liking’ is so easy that it’s almost meaningless. As mentioned above, this will have a positive knock-on effect on all your other metrics anyway.
Accelerating growth and returns
5,000+ followers, you know what you’re doing. Growth should be coming organically, and this is the time to take a big step to break past the plateau. You can’t wait for a piece of content to go viral, the best option for most people is to spend a little ad budget. It doesn’t have to be a lot of money, and a little can go a long way when done correctly.
Do you have to spend money? NO. Does it help? YES.
- Experimenting with copy, visuals and targeting — impressions, click throughs, CPCs, CPAs. With paid advertising, you’re dealing with a whole new set of data points. What’s important here is that you test, test and test some more. No matter what your budget is, you should be spending 20–30% on testing out your creative. Any bit of saving you can make can make a huge difference in the long run (the remaining 70–80% of your budget).
- Measuring returns — Only pay for support if you have a particular reason to (e.g. to increase your following? to drive people to your site?). There are brands out there that are paying because they think they should, but they don’t understand what it’s supposed to do (just in case it helps) — this never works.
- Channel returns — Paid support is often relegated to just SEM/PPC. However, there are many traditional methods that shouldn’t be forgotten (this is where an Excel spreadsheet comes in handy). Posters, PR, advertorials, even billboards still work when done well. What you need to make sure is that you are tracking everything you can (much easier with digital than traditional advertising). Even free cross-collaborations should be measured to see if you are getting returns. Again, what works for one brand doesn’t necessarily work for others. The opposite is also true here.
Paid support can go a long way. Billions of dollars are spent every year for good reason. But it’s not just the big wealthy companies that are the only ones that can. As long as you’re willing to experiment and track results, you should easily be able to tell where to best spend your money — ultimately saving not just money but time as well.
Optimising and staying ahead of the game
Hopefully at this point you’re not working alone, but fret not if you are (it’s still possible). This is the stage where you make all your efforts go further than ever before. Small tweaks here and there can reap massive results.
- Constant iteration — social media is evolving every single day, and you’re going to have to evolve along with it. Your content/ads may be working now but there’s no guarantee that it’s going to work a week from now. It makes no sense to rely on one platform (say Instagram) or ad network (say Google Ads) when tomorrow the platforms may change their algorithms or competition becomes too high. Keep analysing and change/evolve as needed.
- Looking outside your industry — you should already know what your competition is doing. What’s more interesting is what neighbouring fields are doing (say if you’re in electronics, you should look at the beauty and media fields) — often the best tactics can be adapted for your own use, but you have to stay up to date and willing to try.
Staying ahead of the game means that you need to be constantly reading and evolving. Measuring is important across the entire process (you never know that data is going to tell you or even what data is going to be meaningful further down the line). If someone says they have it all figure it, question it immediately (they probably don’t, or have out-dated case studies from years ago).