Your customers today are pre-educated. Various studies estimate that up to sixty percent of the buying decision is made before customers engage with sales. (If you’re a marketer, then you will already be familiar with the statistics). This gives us marketers an interesting challenge. On the one hand, more informed prospects make for better leads, but these pre-educated buyers carry risk. They are more easily able to switch providers if their needs and expectations aren’t met. They are also more likely to communicate any dissatisfaction which can quickly become a viral nightmare. This is the reality of the new customer environment.
So we know the pressure is on and we know we have to up our game as marketers to satisfy this new informed and powerful customer! But how? That’s where data-driven marketing comes into its own. It pivots the marketing department from ‘gut feel’ to an outcome focused approach. Personally, I love this new approach where marketers can create effective, efficient campaigns based on insights gleaned from analytic models. It also reduces customer fatigue resulting from over-targeting with irrelevant and unwanted communications.
Just to be clear, when I say data-driven marketing, I mean the many different sources and tools; structured and unstructured data points, including , sales & marketing automation tools, social marketing, transactional data and various other sources to determine the targeting, timing, channel preference and content of your marketing promotions. 
One of the most powerful methodologies behind a data-driven marketing strategy is predictive analytics.
Predictive analytics lets you understand how behavioural patterns at different stages of the customer creation process do or do not influence each other.
A great example (and live example as we have recently had one installed in the office!) is the smart vending machine. The smart vending machine gives you a completely customised, personalised experience. I swipe my badge and the machine has a history of my last purchase. It links to my Facebook account with the prompted option of whether I’d like to purchase anything for my friends. It also offers me alternative options based on my history of food and drink preferences. And just so I’m never caught short without cash, I can transfer money from my bank into the machine, and store funds in my account. The amount of my purchase is then automatically deducted from my funds.
For the vending machine manufacturer, the data provides an aggregate view of the best-selling items, most profitable vending machines and net profit – all based on location. For example, a specific soft drink may be a big seller in one location but less so in another. This helps with customising inventory stock by location. And because the data can break down sales by days, it can predict which sales of which products are likely over set time periods, creating sales forecasting at a click of a button.
This is just one example, but there are numerous other examples of predictive analytics transforming marketing mix modelling, enhancing lead scoring and accelerating up-selling and cross-selling opportunities. Predictive analytics applications can have significant commercial implications for marketing on new customer acquisition, optimisation of campaigns, promotions, and loyalty schemes, product promotions by channel preference, product bundles based on consumer behaviour as well as price optimisation. Deciding which application to start with depends on your business model, the state of your marketing data, your analytics maturity level and of course the mission of your marketing team.
Technology is enabling us to do amazing things when it comes to predictive analytics and data-driven marketing. I feel this can only be a positive in better equipping marketers to create customer relationships that are ultimately more rewarding for clients. So Maths really does have a lot to do with it!
Take a look at this free white paper from IDC for more information around this topic.
Sian Smith, Head of Marketing, SAP EMEA
- SAP Managed Tags:
| User | Count |
|---|---|
| 5 | |
| 2 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 | |
| 1 |
