Craig Hannabus, Strategy Director at Rogerwilco, outlines the limitations of audience measurement in marketing and proposes some solutions.
For the longest time, we used Living Standards Measure (LSM) as the rule by which we defined audiences. The assumption was that if you enjoy a certain quality of life with access to certain amenities and luxuries, you fit into a category. It was neat and it was easy. The categorisation helped us figure out which platforms you’d more than likely pay attention to and what kinds of goods and services you could probably afford. But there were some problems.
Firstly, the LSM questionnaire is far too long (29 questions). When questionnaires are too long, boredom sets in and the quality of the answers diminishes. When a respondent sees that the end is in sight, they tend to rush, missing out any nuances that a question may have, and in some cases simply ticking boxes for the sake of.
Secondly, it turns out that LSM doesn’t paint a very accurate picture. A typical South African LSM graph depicts a massive middle class, which we know isn’t true. Socio-Economic Measure (SEM) paints a far more accurate picture, showing a larger population that are living in a mid to low SEM. The SEM data matches more closely our current economic and employment circumstances.
Unfortunately, the transition from LSM to SEM in the marketing world has been somewhat slow. There have been no devastating issues as a result of mostly using LSM based data. Most low SEM and LSM households can be reached by print and television. It seems to make little difference which measure you use.
In the digital marketing context, the transition to SEM has also been slow because we mostly use interest-based targeting. Most digital campaigns are built around what people have shown interest in previously. The connection of SEM (and even LSM) to digital marketing isn’t crystal clear. There’s precious little by way of tools to transfer that information into a digital context and its likely because nobody has seen any value in doing so.
When LSM and SEM data is used, it’s usually just to flesh out an audience or a consumer portrait so that everyone in the room is comfortable that we know who we’re speaking to. It may also play a role is in messaging structures, but even then, these measures provide very little extra to what a brand may already have.
At the end of the day, we’re sitting with two very broad ways of thinking about our audiences: SEM and Interest Based Targeting (IBT). SEM tells me what your life is like, IBT tells me what you’re interested in. I’m left with no idea as to why you may do the things you do or what your challenges are. It’s like trying to buy someone a useful birthday present when all you know about them is that their favourite colour is blue.
The solution is relatively simple. Qualitative data, interviews with current consumers, grass roots deep dives – all of these go a long way to painting a consumer portrait. This will give us the why. It’ll tell us what people are thinking and hopefully feeling. We’ll end up with some granular data that we curated and will apply directly to our brand. Yes, it’ll cost a fortune, but the data will be useful for years to come.
But then a pandemic strikes. People’s economic situations change, their behaviour changes, and their priorities change. Suddenly, your carefully curated qualitative insights aren’t as trustworthy as they once were. Do you spend the money to re-do all the work? The solution lies in first-party data.
The first-party data conversation is for the most part been driven by the clampdowns on Facebook and Google’s privacy when it comes to advertising. But it goes beyond that. If you have a captive audience that you are providing value to and who are consistently engaging with you, you are sitting on a goldmine of insight. You just need to learn how to mine. Unfortunately, if you look for good first-party data strategies online, you’re going to get a lot of articles explaining how to capture data with lead forms and social advertising. It’s great to have the data, but if you don’t have a plan, it’s like having a tank full of petrol and nowhere to go.
Show your audience content and see what they engage with. Show them more content, watch them engage. Iterate over and over until you’ve developed a clear portrait of who your customer is. If you’re not sure how valid this approach is, just look at Facebook and Google, they’ve been doing it for years.
It’s harder than it sounds. Every step of your customer’s experience needs to be mapped out and analysed. Each touchpoint within that journey will require you to understand exactly what your customer is thinking, doing and feeling. If your customer journey isn’t helping your customer think, do and feel positive things, that’s a place you need to focus on. Here’s the bottom-line: LSM and SEM are starting points, IBT is useful to a degree. The gold is first party data, but it isn’t enough to mine the gold, you need to learn how to make it into something useful.
ROGERWILCO
www.rogerwilco.co.za
