The biggest problem every marketer faces is knowing where to allocate their marketing budget, and how much to spend.
And as marketing gets ever more complex, with even more choices of channel, especially in digital marketing, this problem only gets bigger. For example, how should you allocate your funds across TV, radio, social media, direct mail and daily deals?
So finding an accurate way to measure your campaigns and predict the best possible future outcomes is essential.
However, most modern measurement methods rely primarily on data from cookies, which only gives you a very limited view of your advertising touch points and your customer lifecycle.
And with the third-party cookie gradually being phased out, it’s becoming more and more difficult to accurately attribute and predict the true potential and performance of your campaigns.
How Total Media Attribution solves this problem for you
We’ve developed our own attribution model called Total Media Attribution (TMA). TMA takes a holistic view of your overall business to make smarter decisions and increase marketing efficiency, both online and offline.
The aim of TMA is to understand how media spend affects your sales, and to optimise the allocation of your spend across media.
TMA identifies where your conversions come from, both online and offline, across a variety of different channels. It measures the success of your current campaigns and, more importantly, it simulates conversion returns on future campaigns, giving you valuable insight to inform your marketing strategy and planning.
TMA works across today’s complex offline and online marketing ecosystem, from TV, radio, print and PR to digital – including Facebook ads, Google Ads, programmatic, DOOH, CTV and audio.
Our accuracy of prediction reaches statistical significance (over 95%) and aims to be around 99%. Enabling us to predict exactly where your returns will come from, and how you should weight your spending across your entire marketing campaign.
We can even factor in seasonal variations and geographic preferences. So if people in Germany prefer print, we’ll upweight that. And if people in the US like video, our predictions will factor that in.
We can show you what returns your campaigns will get with each balance of media – and what will happen if you do nothing at all via a particular channel.
How do we measure Total Media Attribution?
So how exactly does TMA work? We use an advanced statistical model called the media mix model (MMM) using the Bayesian approach.
Put simply, the media mix model assumes that:
- The impact of marketing isn’t immediate, there’s a delay.
- The impact of marketing isn’t global, it’s regional.
- The impact of marketing isn’t consistent, channels perform differently.
- The impact of marketing isn’t easy to capture, it’s diverse.
We look at inputs such as conversions per week (i.e. number of orders or amount of revenue) and marketing spend per channel, as well as macro and micro factors. This builds a statistical model that predicts revenue for each channel.
This then provides us with invaluable insights such as:
- Whether you will reach your revenue targets using this approach.
- The marketing impact per individual product.
- Your bespoke optimal marketing mix.
This enables you to make informed decisions about where to spend your marketing budget, and how much to weight to each individual channel.
Six key benefits of using our Total Media Attribution model
There are many benefits to using TMA. Here are six of the key advantages over MMM:
- It can be used on an in-depth regional level, and we don’t just look at immediate returns; we model how returns change over time (and lag response).
- It works online and offline and allows you to balance 13 variables, including spend, base brand and Covid cases.
- Unlike the traditional approach of MMM, we don’t need to look back on six years’ worth of data to make our predictions, making it quicker and easier for you to get started.
- TMA can natively integrate macroeconomic and behavioural signals such as global pandemics and regional brand awareness.
- Providing we are respectful of significance scores, TMA can work with any depth of data beyond six months (compared to three years for traditional MMM).
- TMA models regional and geographic features in ways that MMM cannot, such as a behavioural difference between metropolitan versus rural areas.
Ultimately, our modelling creates a coefficient (a constant quantity) for each channel, representing the number of customers acquired for the money spent.
So you can finally solve that age-old marketing problem of truly knowing where to allocate your marketing budget and how much to spend, and maximise your results.