March 19, 2012
Predictive analytics can produce an average ROI of 145%, compared to an average ROI of 89% for processes which did not incorporate analytics. It does this by streamlining the marketing process, segmenting customers into clusters and ranking them in order of benefit to the company. This means that marketing campaigns are targeted and deliver a higher response rate.
Analytics can also be employed to predict the life span of a customer with a business, so that measures can be employed to extend the life of the customer’s relationship with the business and thus increasing revenue. In a tough economic climate, every little helps.
The process by which analytics helps your business is through the Cross Industry Standard Process for Data Mining, or CRISP DM
To get the best results from data, the CRISP DM process should be followed. It starts with the Business Understanding stage – what is the desired goal of employing predictive analytics? Is it to discover how many people stop using the services of your company, or is it to improve the effectiveness of your marketing campaigns?
The next stage is Data Understanding – what is the quality of the data like? Is the data you have suitable for the goals of your business? If not, then you’ll either have to reassess the goal or collect new data. The next stage is to prepare the data for modelling – cleaning up the dataset and making sure that the cases match the designated variables.
The next stage is the modelling. The possibilities here are only limited by the data. For example, you could employ a Recency, Frequency, Monetary value (RFM) model to find out who your best customers are. Or you could employ propensity modelling to predict who will respond to a marketing campaign. These tools, and many others, allow marketing to be targeted, thus cutting down costs and driving up response rates – increasing revenue.
The next stage is to evaluate the models and ask the question: Does the result of the modelling fit the goals made at the Business Understanding stage? If not, why not? If the models do fit the goals of the business, the results are deployed into the marketing campaign to create a targeted and personalised marketing experience for both business and customer.
Once the marketing campaign is underway, feedback is fed back into the model to a) judge the success of the campaign and b) make subsequent campaigns more accurate.
Employing predictive analytics in your marketing campaigns and incorporating the practice into your business model will produce of wealth of benefits enabling your business to grow in a very tough economic climate.
Contact Us
« Sarabjit Wilkhu Moving On after 12 years I may be a Geek but I’m a rich one »
