Tackling the problem of measuring digital marketing effectiveness
For an industry that perpetually preaches about breaking down silos, digital tends to neglect the same holistic approach to measuring the effectiveness of marketing spend across organic, paid and social. While it’s reasonably easy (provided you have a good attribution model in place) to monitor how paid media is impacting your brand, overall marketing effectiveness is something that has caused head scratching for decades
Back in the pre-digital era of the 80s – before Click Consult, before Google, before the internet, marketing professors Thomas Bonoma and Bruce Clark stated that:
[Perhaps] no other concept in marketing’s short history has proven as stubbornly resistant to conceptualisation, definition, or application as that of marketing performance[.]
While both marketing and the internet have aged somewhat in the meantime, the statement remains largely true. While digital advertising has permitted marketers a far better access to ROI data than Bonoma and Clark would have dreamed of in 1988, there is – especially as Google continues to blur their keyword data – a sense that another voice of marketing past, John Wannamaker, may still have a point when he said ‘[half] the money I spend on advertising is wasted; the trouble is I don’t know which half.’
The truth is that marketing produces both tangible and intangible effects, or as a more recent marketing paper by Jagdish N. Sheth et al put it:
[History] impacts marketing effectiveness – expenditures tend to be accounted for annually, whereas the influence of those expenditures is cumulative, thus a change in sales volume in one year could be the residual echo from previous years activity rather than current activity.
So, what can brands do to ascertain the impact of their marketing – especially when, contrary to the findings of every academic study, marketing budgets still tend to be the first cut during times of economic uncertainty and recession?
In his 2003 text ‘Marketing and the Bottom Line’, Professor Tim Ambler noted that there are five stages of how a company handles the subject of marketing effectiveness:
- The company is unaware of the issue of not measuring marketing effectiveness
- Assessment is introduced but only in financial terms
- Using financial measures alone are recognised as inadequate and a multitude of nonfinancial measures are introduced
- The company develops market focus and the assessment measures used are streamlined to give a single coherent view of the market
- A scientific method of assessment is adopted using a database of past and current metrics, derivatives and diagnostics to produce a shortlist of sensitive and predictive metrics
There will doubtless be one or two steps that will be familiar to marketers, however – and companies tend to vacillate between steps 2 and 3 with few seeming to progress further, and this is what leads to the undervaluing of digital marketing techniques like SEO and the over-reliance on fads like partnerships with the Insta-famous. It’s also why paid media has been a consistent earner for Google regardless of their reliance on black box technology – there are immediate, measurable, financial impacts of paid search.
One of the best descriptions of the situation we still find ourselves in comes from a paper by Brooks and Symkin, who wrote (a decade ago):
To quote Clark (2000, p21), “Clearly managers are capable of assessing multiple dimensions regarding performance. The question, then, is whether they are assessing the right dimensions for their business.”
Marketing academics do not always help. By couching their views in complex technical terms they can easily put off practitioners from considering new measurement options – consider this description for example, “a principal components multinomial logit regression model for estimating the Markov brand-switching matrix” (Rust et al., 2004b, p123) – one that would perhaps be immediately consigned to the rapidly growing ‘too difficult’ pile that exists on every marketing managers desk.
The same paper, however, also offers six points as a potential solution to the difficulty of measuring marketing effectiveness – specifically for SMEs, for which it can be especially difficult:
- Work within whatever historic information the company has (financial, anecdotal, …) but authenticate elements with customer/partner surveys and competitor reviews where possible.
- Consider as many aspects of the full marketing process as are reasonably practical within the confines of the SME’s current situation. As a minimum this might need to be restricted to Segmentation, Targeting and Position and the marketing mix alone, especially when the company culture does not consider anything other than promotion to be ‘marketing’ (Siu and Kirby 1998). Broader elements of the marketing process can be considered in future iterations after the SME has ‘learnt’ to capture relevant information and understand the motives and implications of such.
- Create a jargon-free method of communication that the practitioner and company can use to rationally discuss/decide on marketing matters.
- Use visual indicators to deliver a marketing effectiveness measure (or maybe just a marketing mix effectiveness indicator) that the company can understand and accord with. This indicator will most likely be an aggregate of a number of marketing measurement metrics, the constituents of which may not need to be exposed to the company management.
- Create an agreed activity plan for the marketing areas that will improve future effectiveness along with easily understood measurement criteria to measure that improvement.
- Try to create a company culture that readily captures and retains base data that is useful to future marketing effectiveness measurement.
This is the approach that we’ve been attempting to bring in at Click Consult internally over the last few years, so we’ll go through these and see if we can offer some guidance on how to accomplish this for your digital marketing.
1. Work within whatever historic information the company has (financial, anecdotal, …) but authenticate elements with customer/partner surveys and competitor reviews where possible.
One thing digital does well is historical data – provided you’ve had an Analytics tracker attached to your site – be it Adobe, Google or any other, you will have plenty of data you can use. The important and overlooked aspect of this is the impact that historical competitor reviews and customer surveys can have on determining the effectiveness of marketing.
So, while we can (and do) advise on the importance of analytics and collecting data through analytics, I’m also going to suggest making client/customer surveys a part of what is often referred to as the ‘onboarding’ process – but which is essentially post purchase, or post contract. Ask why they chose you for their purchase or service provision and follow up on that at renewal stage – or a version of it for each additional purchase.
While it may not be quantitative data, this kind of assessment is important qualitative information that can be used to assess the performance of your marketing over time.
2. Consider as many aspects of the full marketing process as are reasonably practical within the confines of the SME’s current situation.
While it is always tempting to see digital marketing’s value as immediate uplift in sales, profit or visibility, it should also be assessed whether your ability to reach your target demographic has improved, whether there has been movement in qualitative measures such as perceived quality, brand awareness and share of voice (on social media, or external articles, for example). It can be reasonably easy to measure the performance of a content marketing campaign, for example, using utm tracking and traffic from specific campaigns, but what is the uplift in general citations of the brand beyond your campaigns?
3 Create a jargon-free method of communication that the practitioner and company can use to rationally discuss/decide on marketing matters.
This can be harder than it sounds – we develop jargon as a language of shortcuts. While jargon has become almost synonymous with what Orwell referred to as duckspeak – a thoughtless kind of waffling – at its core, jargon is a vocabulary which allows professionals within a field to communicate complex ideas in shorthand. The main issue with jargon is that it spills over into conversations outside of the field and it’s this that causes issues with communication.
For that reason, it’s important to address (and address early) the creation of a common vocabulary which can be understood by all stakeholders and to use that common vocabulary whenever issues of measurement, planning and strategy are presented for use outside of the marketing department. You can have the best measurement methodology in your field, but if it can’t be understood by decisionmakers, it’s useless.
4. Use visual indicators to deliver a marketing effectiveness measure (or maybe just a marketing mix effectiveness indicator) that the company can understand and accord with.
Similar to point three, this is about ensuring that there is agreement in place as to what effectiveness looks like (taking into account that it should not just be ‘more revenue’) and to stick to a practice of establishing these in advance of measurement – even if this is an amalgamation of more detailed measurements. The important thing is not comprehensiveness, but comprehension.
5. Create an agreed activity plan for the marketing areas that will improve future effectiveness along with easily understood measurement criteria to measure that improvement.
Again, this ties in with the previous two points – and is the reason we have spent so long over the last couple of years building custom Data Studio dashboards for client reporting. By agreeing not just campaigns and activities, but more holistic measures of success and more comprehensive reporting. This is in line with a statement that I heard somewhere but couldn’t attribute (apologies) which is that top level results – the stuff important to everyone – should be visible at a glance, while those with a deeper knowledge should have the ability to analyse the report at their own level of understanding.
6. Try to create a company culture that readily captures and retains base data that is useful to future marketing effectiveness measurement.
This maybe the most important for ongoing development of measurement – and requires ‘buy-in’ from other departments – from sales in asking the ‘where did you hear about us’ question, from account managers who will need to oversee existing client surveys (or your web development and email team who will need to build a method of surveying consumers), but also from marketers who need to move beyond monthly or quarterly reports to ongoing data collection and a greater attention to historic performance and the impact of historic activity on present performance – however, it also needs the support at c-suite level for the additional resources that that will take and the gains that can be achieved from proper measurement.