Free Subscription

  • Access 15 free news articles each month


Try one month for $5
  • Unlimited access to news,insights and opinions
  • Quarterly and weekly magazines
  • Independent research reports and forecasts
  • Quarterly webinars with industry experts
  • Q&A with retail leaders
  • Career advice
  • Exclusive Masterclass access. Part of Retail Week 2021

Mmm, it’s the sweet smell of advertising success

(Source: Supplied)

Am I the only one who’s suddenly doing a lot more shopping at 8pm? And I swear I’ve never been more susceptible to a digital promo in my life. Twenty-per-cent off at my local Thai restaurant on a Friday night? I’ll have the extra pad see ew, thanks.

Since early 2020, like everyone, I’ve been online a lot more. I’m seeing greater quantities of online advertising and – especially if they’re good ads – feel motivated to purchase. 

It turns out I’m not the only one. Annalect’s latest Marketing Mix Modelling (MMM) analysis1 took a deep dive into the world of Aussie retail and it’s borne out a lot of the behaviour I describe above. 

Here’s the first interesting finding. Locally, the retail industry sustained substantial growth throughout the first pandemic affected year (2020-2021). Yes, you read that right. 

For the year to June 2021, consumer retail spending grew by 9.1 per cent. This is a remarkable figure when you consider that for the four previous year-on-year periods it never exceeded 3.1-per-cent growth2.

From this we can glean a few things. Firstly, a lack of physical access to brick-and-mortar retail outlets has been no barrier to growth whatsoever. Secondly, it speaks to a certain level of digital preparedness from the retail industry when the pandemic struck – or at least the ability to change tactics quickly. 

Some of the lift in retail sales we can also put down to a siege mentality in the face of uncertainty – remember when you couldn’t buy toilet paper for love nor money? Or those long, restless days in lockdown when we turned our attention to our homes, driving record spending on furnishings, technology and renovations. 

But most significantly, I think it shows how quickly our new retail habits have become second nature, with 9.4 per cent year-on-year growth in retail consumer spending to March 20223. So let’s dig a little deeper.

What is Marketing Mix Modelling anyway?

If you work in retail you may be familiar with MMM as a concept, but chances are the finer points are still a bit murky. And while it’s by no means a new approach, it is growing in importance and usefulness. In an era of enhanced privacy concerns and decreasing availability of conversion event data, MMM remains resolutely ‘signal proof’. 

Its recent evolution (including being more agile, automated, and machine learning-driven) has also enabled marketers to answer more questions at a time of increased scrutiny on the CMO’s bottom line. 

Put simply, MMM is a statistical analysis of sales and marketing data that helps measure the effectiveness of marketing tactics. Commonly it is undertaken at an individual brand level, but with collaboration can be extended into a meta study of a particular industry and may look at years worth of historical data from a wide array of brands.

We were fortunate to receive enough industry collaboration to undertake a comprehensive meta study of the retail industry incorporating home, health and beauty, and consumables. The categories covered campaigns that totalled in excess of $170 million.

MMM studies will typically measure marketing effectiveness against a ‘base’ figure, which is defined as the number of sales that would occur naturally if no other action was taken. Where it gets interesting is when we look at ratcheting up retail spend on, say, a particular media channel, and then assessing the effects.

This is precisely what Annalect has done. With data that covers the extent of the pandemic and back to 2017, our MMM is a deep dive into how Australian retail has evolved in recent years.

New channels for a new era

Looking at the retail landscape, on average, five dominant channels – TV, display, search engine marketing, radio and Meta (Facebook) – accounted for 84 per cent of media contribution and 82 per cent of media-generated sales. Of these channels, TV attracts by far the largest media investment, at 37 per cent.

We know, however, that since 2020 digital channels have been more important for both businesses and consumers. This, and its general saturation, goes some way to illustrating why TV tends not to fare as well when additional spend is modelled into baseline investment. 

Across the MMM study, an additional 10 per cent of budget injected into TV showed an 8.1-per-cent lift in return on investment. Conversely, digital mediums such as Meta and Out of Home, which attract a lower level of investment overall, tend to perform better from an ROI perspective. An additional 10 per cent of budget added to baseline Meta campaign spend yields a lift of 16.8 per cent, while Out of Home sees a lift of 12.8 per cent.

Peeling back more layers, one reason for Meta’s strong performance, in particular, is that most brands analysed on the platform were getting the advertising basics right. 

That is to say, they had sufficient reach, appropriate frequency, and adopted known Meta creative and media best practices, such as six or more placements (for example, on Facebook Feed or Instagram Stories). The clear lesson: get the essentials right if you want to catch people’s attention and drive them to purchase.

Our overall findings from the MMM suggest that a broad channel mix is still very important for retail marketers. But more than ever, an increased focus on digital channels and ensuring you adhere to a few key advertising basics look to be the best way to woo your customers. Now, where was that promo for my local Thai restaurant again?

  • About the author: Nerida Jenkins is marketing science lead at Annalect.


  1. Meta-funded Marketing Mix Modelling analysis by Annalect
  2. ABS 8501.0 Retail Trade, Australia. Table 12. Retail Turnover, State by Industry Subgroup, Seasonally adjusted
  3. ABS 8501.0 Retail Trade, Australia, Retail Turnover, Seasonally Adjusted, 4th May 2022 release