If you’ve got a robotic vacuum, it might be collecting more than just dust. It was recently reported that high-end models of the Roomba, made by the company iRobot, gather data about the location of your furniture and walls as they move around your home.
This led to concerns that these maps might then be sold on to other companies without a customer’s knowledge. iRobot’s chief executive later clarified by saying they were hoping to share the maps free with customer consent, but it does raise some fascinating questions about why this data was being collected and how it might be used by other organisations.
Marketers vacuum up data
It’s not the first time that innovative methods have been used to source information about customers. Primary data from surveys and interviews is useful, but there is a limit to how much we can rely on that. Deeper insights often come from a customer’s behaviour (such as sales and financial transactions), or information about their circumstances that wouldn’t be collected on a questionnaire. That data might be generated internally from a company’s own records, obtained from a commercial partner, or gathered from widely available demographic datasets.
Take Denmark’s largest bookstore, for example: the company managed to track the website activity of individual customers and used it to send automated emails related to what a customer had shown interest in. Or satellite media provider DIRECTV, which used data from the USPS’s list of recent movers to target those who might be receptive to a new provider — and got double-digit conversion rate lifts as a result.
But why is this customer data of such value to marketers? As the previous examples suggest, it’s all about tailoring our campaigns to make them more effective. Increased personalisation is proven to boost engagement and ROI, and that includes when and how a message is delivered, as well as the message itself. To know how to market effectively, we need to gather lots of information about our customers and and how they might react to our efforts.
But even then, that mountain of data doesn’t really tell us much on its own; to make sense of it and guide future actions we need to use a variety of different analytic techniques. For marketing purposes, we can divide those techniques into three different categories of analytics:
- Descriptive
- Predictive
- Prescriptive
Types of analytics
Descriptive analytics is all about looking at past events to discover what happened, using visualisations and finding those hidden insights. We might analyse past marketing campaigns to see how they performed, reporting metrics such as response rate and conversion rate to determine overall success. One tool many might be familiar with is Google Analytics, which allows companies to see what effect promotional campaigns might be having on traffic to their websites.
Predictive analytics, on the other hand, is all about looking to the future. As the name suggests, it involves predicting what is likely to happen; building on the data studied in descriptive analysis, we can identify patterns and trends right down to the individual customer, and produce forecasts to inform our marketing efforts. Knowing what a customer might want next allows us to target them better — just think about Amazon’s ‘Customers who bought this also bought…’ section. Another famous example is – suitably – the retailer Target, which found a way to predict if a customer was pregnant based on their purchases.
Prescriptive analytics is less widely employed than the previous two methods, but can actually deliver even more value. This is because it goes beyond just predictions to recommending actions based on those predictions and showing the consequences of those actions. Using a combination of big data and techniques such as machine learning, we can produce simulations that inform the decision-maker. Boots, a UK-based pharmacy retailer, is using prescriptive analytics to inform their choice of store layout, including how the placement of product promotions will affect sales.
As we can see, employing marketing analytics can be hugely beneficial for a business. Going back to our Roomba example, that data collected by the robot vacuum (about room size, furniture placement etc.), could be used to predict what an iRobot customer might be interested in buying next, to produce highly targeted promotions. It might inform a prescriptive analytics simulation about how successful a campaign will be — if you don’t have much available space in your living room, how interested are you going to be in large couches?
The personalisation that can be achieved through analytics is a key theme when we look at data driven marketing success stories. Just take a look at Australian bank Westpac, which managed to grow customer engagement from 1% to 25% in just nine months, by analysing customer behaviour to better match them with new programs and offerings. Or how about Amex, which says that by using predictive modelling it can now identify 24% of Australian accounts that will close within four months. Identifying those customers that are about to churn is half the battle when it comes to retention.
Recognition of these techniques is helping to fuel increased marketing budgets, and is reflected in the fast-growing digital advertising market — companies can have more confidence in these campaigns if they are targeted correctly. In fact, data-driven marketing was the top strategic priority for 53% of organisations in 2016, with marketers spending more and more of their budget on analytics. Specifically, the use of prescriptive analytics is expected to become more widespread, with Gartner estimating that 35% of organisations will employ it in some form by 2020.
To learn more about how marketers use data analytics to reach new and existing customers, check out the original infographic created by Villanova University’s Online Master of Science in Analytics program.
However, there are question marks over whether we, as an industry, are ready to take advantage of these new technologies. Data analysis for marketing initiatives is one of the top career paths for data scientists, but according to a 2017 report, 40% of marketers believe a lack of data and analytic skills is preventing them from delivering effective CRM strategies. Knowing the importance of data-driven marketing and analytics is one thing, but implementing them is another — it is clear we as marketing professionals can do more when it comes to identifying and recruiting the right talent needed to achieve this.
Have an opinion on this article? Please join in the discussion: the GMA is a community of data driven marketers and YOUR opinion counts.
Read also:
The power of data – data may be the new oil, but it’s only valuable if you have a reliable engine