On running an SMS campaign in Nigeria
First of all, this is more a documentation for myself because it was my first time doing an SMS campaign.
I was saying the other day that I prefer written convos bc I can come back to them later + remember. But that also applies to sharing stuff.
— — (@heykastrah) October 17, 2024
We often share to “teach” or “guide” others, but what if we just share to remember? Thinking about the mentality switch — it makes sense.
The first shocker for me was how different virtually all SMS platforms in Nigeria are. I've seen Recart and had similar expectations, but after checking a bunch of platforms, I realised we were playing a different game.
I'd learn later that it's not the fault of these platforms since there's little to no incentive for the NCC to improve the user experience, so things like analytics isn't all that clear with SMS campaigns. Secondly, it seems we're more about email campaigns in this part, which is another thing that makes me curious.
Is the general consensus that SMS campaigns don't work or are people just not talking about SMS campaigns as much as email campaigns? Someone I spoke with simply said "how many people are doing SMS campaign?" as to why there isn't much zeal to improve the campaign experience.
Thinking of it now, I mostly hear about SMS campaigns during BBN and elections when politicians won't let you rest if your number unfortunately falls into their hands. I remember one Ilorin politician sending SMS to folks and we were all wondering how he/his team got our phone numbers, especially since some of us aren't indigenes.
I saw this tweet from a mutual last year and thought of how interesting it is that in some parts of the world, SMS is the more preferred channel.
Filipinos hate emails 🤣
— Kristiana Rule ✨ (@kfromcopylang) February 16, 2023
The alternative: SMS
This one’s from my favorite eatery. They text right before lunch every other day.
The kicker:
The transaction happens over text too. pic.twitter.com/YINMTeRFzg
Anyways, due to the issues with getting granular analytics, it seemed obvious that you can't simply run an SMS campaign without having other things set-up. It's like an ecosystem — first, you need a Meta Pixel to retarget folks who come via SMS, and second, you need an email campaign/flow for people who complete actions you're optimising for on your website. Luckily, we already had these covered, so I wasn't totally starting from scratch.
I'll admit it took some hours for me to see this — which is really why I'm documenting it here — but if the goal of analytics is to know the performance of campaigns and the end goal isn't just a link click, then it means you can do all it takes to measure at the website level. My initial mistake in thinking was being stuck on measuring the effectiveness of a channel within the platform — you know how you can see analytics on your email platform.
But it's also possible to measure the effectiveness of A by measuring the activities that occur in B and C. For example, if you send an SMS to 6,000 people and only 1,250 land on your website, it's obvious there's an issue with the messaging. If 3,000 people land on your website, you know for certain that the messaging is working.
You can then convert from the 3,000 with other stuff you've put in place. When someone creates an account, they get your welcome flow, and when they make a purchase, they get your post-purchase flow. Your Meta Pixel also then becomes useful in targeting these 3,000 people on Instagram and Facebook, for example.
Something that's definitely critical are UTM links. One thing I thought about was how to know which message is performing best — you can't improve what you don't measure. I may have been under the rock with this, but being a cheapskate who doesn't want to upgrade on Bitly led me to Google's Campaign URL Builder, which is totally free to use. Funny thing is that you can then use Bitly to shorten the link.
With more than one messaging angle, having more UTM links that are differentiated by the 'content' helps to see how each message is performing. And since we use PostHog, enabling their 'URL parameters to event properties' app was just the obvious next step. Now it's easier to segment people performing website actions via the UTM links based on set parameters.
I definitely wasn't expecting to enjoy this side of things but I was awake at 4AM, so it was really fun to arrive at a dashboard that shows how each message is performing in comparison to other messaging angles.
Found the thread below while writing this and it's even more reason to have an ecosystem play for SMS campaigns.
Open rates for SMS marketing are misleading—it's best to avoid using them as your defining metric.
— Black Jaguar (@A_Feranmi) June 18, 2021
Some reports suggest that SMS has a 99% open rate. But the majority of those opens might be users: