by Anthony Pey, Head of Marketing Effectiveness at Medialab
In today’s fast-changing industry, accurately measuring how audiences respond to marketing campaigns, and adapting budgets accordingly is vital for staying competitive. This is particularly important for multi-channel advertisers because the way audiences respond to campaigns is always nuanced – often audiences don’t respond immediately, or they use a series of channels or even devices along their path to conversion. This means the effect of media campaigns can often be under-represented in linear tracking present in cookie-based, click-based or other direct response means.
There are a wide range of measurement techniques that advertisers and brands can use to understand the effectiveness of their marketing efforts. The use of traditional tracked attribution versus econometric modelling is a subject of ongoing debate. Where attribution has typically been used to better understand the individual customer behaviour and journey, econometrics takes a deeper dive into the true incremental uplift of individual channels, alongside their longer-term contribution and impact of any wider macro factors. Specialist measurement techniques have been designed to capture the full effect of media and evidence the correct view of media incrementality and accountability, thereby showing the full effect of marketing and how it can be used as a force for growth. Effectiveness expert Les Binet is typically referenced by many when comparing the two, having previously proclaimed that “Attribution – quickly and cheaply – will give you an answer that is precise and wrong. Econometrics – slowly, laboriously and expensively – will give you an answer that is right.”
So how can utilising differing metrics and econometric modelling help advertisers and brands adopt a more agile and holistic approach to measurement?
Employing multiple metrics to better understand marketing impact
In recent years, marketing metrics have shifted from merely measuring likes, followers and clicks and performance from last-touch results, for these can arguably skew the marketing impact on bottom-line results. Instead, brands are looking to create a ‘full funnel’ understanding of campaign success. This has been powered by the evolution of data-driven techniques which are empowering marketers to react swiftly to shifting customer needs vs. market behaviour, compounded by increasingly new and diverse ways to reach and influence consumers.
As data goes through various steps, from collection to measurement, analytics, and optimisation, using the right combination of metrics can generate valuable insights for advertisers and brands. A blend of multiple metrics, both fast and slow, from different sources can help better understand marketing impact and help demonstrate the full return on investment (ROI).
To effectively measure marketing impact and performance, advertisers and brands can consider tracking a mix of lead generation, revenue and sales, brand awareness, and engagement. Tracking each stage of the customer journey can help brands and advertisers better understand which channels are driving these performance metrics, be it social media or even email marketing.
Measuring traffic, links, time-on-site, click through rates alongside conversions and revenue can allow advertisers and brands to holistically view this crucial data. By synthesising the multiple touchpoints that customers interact with across the marketing and sales funnels, they can view a unified picture of their marketing impact and performance.
Achieving an ‘edge’ through econometric modelling
Given the reality of dealing with inflation, brands and advertisers often need to react quickly and with agility when it comes to optimising their media spend.
Econometrics – otherwise known as Marketing Mix Modelling (MMM) – is the gold standard in terms of measurement techniques. It can be used to help determine which factors influence volume-based KPI’s and by how much. The technique can disentangle the impacts of two or more factors that are changing at the same time, such as different pieces of activity running concurrently (i.e., TV, print and digital). Econometrics can isolate the uplift in sales driven by media activity, far beyond what can be seen in the trackable universe, showing both the immediate response and non-linear responses that occur in days/weeks after the activity is seen. It shows the contribution and relative effectiveness across all media channels, measured through the same technique for direct comparison i.e., TV, print, display, paid social, paid search, CRM and more. Econometrics is robust, and well-recognised within the industry as the only way to demonstrate the true effect and value of media activity in terms of ROI.
Often praised as a tool which eliminates the guess work in advertising, guiding brands on where ad spend is best allocated, econometrics equip advertisers with a holistic outlook for quantifying the sales impact and ROI of marketing activities. With econometrics at the core, brands and advertisers can make the right decisions about where to allocate budget, and other analysis points such as price and promotion, product strategy, salesforce or campaign optimisation, integrating the customer journey into the same framework.
At Medialab our proprietary marketing intelligence platform, Apollo, can help speed up the time it takes for econometric modelling as it improves access to organised data, meaning our clients have regular and timely access to the correct measures of success that econometrics can provide.
When is the right time to deploy and invest in an econometric model?
The decision to utilise an econometric model can typically depend on the size and complexity of a business, its availability of data, and the goals of its marketing efforts.
A smaller business, for example, with a limited budget but the ambition to learn and grow, may want to deploy an econometric model to understand the main factors that influence their overall KPIs, i.e. total leads, or total sales volumes. They can then understand the contribution of advertising for each of their marking channels, beyond tracked responses to explore which channels to re-prioritise during the upcoming campaign.
At Medialab, we recently worked with a smaller charity to re-optimise its limited budget. This was based on the understanding that its winter TV campaign was under-represented by 60% of its true value, and paid social had a much more acceptable cost per acquisition, but importantly only within certain investment headroom.
For larger businesses or organisations, more complex versions of econometric modelling frameworks can be utilised, looking at different KPIs, such as profitability of driving leads to different sales channels, products or even softer metrics such as improving brand health scores and market share.
In this instance, Medialab’s modelling identified profit gains that dwarfed the modelling investment cost by almost 95%. This was done by producing a measurement and optimisation framework examining TV peak vs. daytime investment, alongside how the laydown would support engagement through to website, and leads captured through increased digital investment.
Experienced econometric and measurement providers will often develop and tailor frameworks that are scaled against media investment (normally a low % of the overall media budget), balancing the granularity of models (alongside cost and complexity) against business questions such an exercise seeks to answer.
There is often a disparity between providers and pre-made solutions, good outputs are indicative of good data inputs, handled by industry experts that relay the insights and recommendations in clear and concise ways. Brands should consult their media agencies, trusted partners or in-house specialists in the field.
In all, with the growing need for agility in marketing measurement, a holistic approach can help evaluate effectiveness. Employing multiple metrics across different sources, coupled with econometrics and attribution can help brands and advertisers form a coherent story around the impact of their marketing mix. This should be created as a durable framework, designed and scaled around media budget, enabling future-proofing measurement capabilities. It should also be delivered by experienced specialists that understand how best to drive sales and overall business performance through model-based learnings.
