Jul 21 '14
At EC Towers we don’t usually subscribe to complicated management consultancy reports… or find them very interesting. However, global management consultancy firm, McKinsey & Company, recently conducted a review of more than 400 diverse client engagements (ie. marketing things people did) from the past eight years. They were specifically looking at the ways that people monitored and measured the success of their marketing and the impact this could have on the bottom line. As a result of this they found that an analytics approach incorporating more than one method could free up around 15-20% of marketing spending. That equates to almost $200 billion worldwide that could be reinvested or filtered down to the bottom line of each business.
But, the fact is… with a plethora of strategies and data methods flooding the analytical market at the moment, it’s easy to become paralysed with indecision resulting as to what to do for the best. This report suggests that you need not be so worried about it when in fact combining different methods would yield a much more accurate and in-depth range of data.
Businesses have traditionally allocated their budgets based on analysing the results of previous years; what line, product or area of the business performed well and why? Although this approach can provide useful insights, McKinsey & Company warns that it can create a ‘beauty contest’ scenario that only rewards the ‘coolest’ proposal or the departments that shout the loudest. A more useful approach is to measure proposals based on their economic value and return instead, allowing marketers to create equal comparisons and combine these with other measurement factors such as baseline spending, prior commitments and thresholds for certain media channels.
Elephant Creative blogger Nikki Bruce takes a look at the findings of the report and how you can use.
Using analytical methods to make better decisions
Whilst the data provided from new analytical methods is substantially more in-depth and relevant than in previous decades, business judgement and experience is still needed to make informed marketing decisions. McKinsey & Company urge businesses to begin by identifying the best analytical approaches for their needs. The three main areas of analytical approach are detailed below.
1. Non-direct marketing choices
These are advanced analytical approaches that assess the short term and long term effects of marketing activities. The most popular of these approaches is Marketing Mix Modelling (MMM), a practice that uses big data to determine the effectiveness of spending channel by channel.
MMM links marketing activities to other drivers of sales, particularly looking at external variables such as seasonality, competitor activities and promotional activities. This method reveals the longitudinal effects of marketing activities such as changes in individuals and segments over time, as well as interactional effects such as the differences amongst offline and online activities and social media efforts.
Marketing mix modelling is useful for long-range strategic purposes and near-term practical planning but also has limitations to its effectiveness as an analytical method. MMM requires high-quality data on sales and marketing activities spanning a period of years, along with users who possess a considerable depth of econometric knowledge to understand the models and scenario planning tools that are part of the approach. Marketing mix modelling is also unable to measure activities that cause little change over time and they cannot measure the long term effects of any one touch point. For example, the long term effects of a new mobile app or social media feed on sales figures.
2. Heuristics (experience based) approaches
Heuristic approaches can be used as a faster but more approximate way of analysing data from marketing activities and one of the most widely used heuristic approaches is RCQ or Reach, Cost, Quality.
RCQ divides each touch point into the component parts of; the number of target customers reached, the cost per unique touch and the quality of the resulting engagement. The RCQ method is able to do this using a combination of data and structured judgement which is why it is seen as a less reliable form of data analysis.
RCQ is generally used when other methods are not viable because of a lack of sufficient data or when the rate of spending is relatively constant throughout the whole year and is spent on ‘always-on’ media where marginal investment effects are harder to isolate and examine.
There are many advantages to using the RCQ approach including its easy to use format (can be executed using nothing more than an excel spread sheet) and its ability to bring all touch points back to the same units of measurement, allowing them to be easily compared. However, RCQ analysis is held back by being heavily reliant on assumptions and difficult to implement – calibrating the value of each touch point can be difficult due to their varied nature.
3. Emerging approaches
Attribution models of marketing analysis are becoming more widely used to aid business leaders in making informed choices when allocating and assessing marketing spending. Attribution modelling refers to using a set of rules or algorithms that govern how credit for traffic to sales conversions is assigned to online touch points i.e. email campaigns, online advertisements, social media campaigns or a company’s website. These credits then help marketers to evaluate the success of different online investment activities and see how they are driving sales.
Most of the scoring methods used in attribution models of analysis take a rules based approach when assigning credits. For example, the touch point that receives the last touch or click before a conversion is given 100% of the credit.
These models of analysis are still relatively new and are being improved continually. Newer methods use statistical modelling, regression techniques and more sophisticated algorithms to provide a more analytical approach than previous practices.
As do all methods of analysis, attribution models do have limits – they depend on data collected by cookies which significantly limits the richness of the data and makes it hard to accurately attribute the importance of each online touch point.
Integration of approaches
McKinsey & Company goes on to explain how understanding each method of analysis and integrating them into a comprehensive strategy can provide the best MROI insights. Using a mixture of approaches reduces the biases inherent in any one MROI method and provides a more accurate set of data. Below is a breakdown of how the various methods can be used together;
Often, TV, digital, print and radio marketing activities account for 80% of marketing spending. These activities generate audience measurement data that can be tracked longitudinally so using a marketing mix modelling approach will yield the best results.
Data gained from digital activities can also be further refined using attribution modelling; this will help decision makers to pinpoint online activities that are most likely to generate the highest number of conversions.
Finally, the heuristic analysis method RCQ can be used to monitor the results of the remaining 20% of spending which often goes towards sponsorship or out-of-home advertisements. The results of these campaigns are harder to scientifically prove and marketers must rely on informed judgement to determine them.
Once a good analytical program has been established, marketing teams can use this data to fine tune the marketing mix, creating the right balance of digital and traditional marketing activities.
Centre your analytical approach
Many organisations outsource their analytical requirements or have a dedicated internal team to analyse data and produce reports. McKinsey & Company advise against this segmentation and encourage businesses to have an analytical approach at the centre of their organisation.
When data is analysed independently by an internal team or outside agency, it is often found that departments in other areas of the businesses are unwilling to effectively use the data produced as they do not trust or understand it in enough detail. Instead, marketers must work closely with data scientists, marketing researchers and digital analysts to question assumptions, formulate hypotheses and fine-tune math to perfect their marketing and analytic strategies.
It is also advisable to cultivate ‘translators’ who both understand data analysis and “speak business” to bring the creative and analytical sides of the business together. This enables organisations to react swiftly to the results of their marketing activities, tweaking spending allocations and campaigns in response to the data being received.
The report from McKinsey & Company highlights the changing nature and importance of marketing analysis. The report emphasises that an integrated analytical approach is the key to unlocking meaningful insights and driving growth in an economy that increasingly demands proven MROI results from business leaders.