Close this search box.

E-mail marketing & data mining, or how to develop a real competitive advantage

aude-demoulin-carrePurchase history, browsing data, e-mail campaign results or other declarative data... Scattered and multiplied, the digital data stored by companies on their customers and prospects is now voluminous and constantly increasing. But if the collection of information is nowadays unavoidable, it loses all interest without the analysis of the latter. Under cover of "pompous" terminologies, data mining definitely deserves to be looked at, especially in a spirit of complementarity with e-mail marketing. Point by Aude Demoulin, Consulting Manager at Dolist.

Data mining, behavioral scoring, predictive analysis, business intelligence... Complex names that give a rather elitist impression. And yet, this analysis technique plays a simple - but essential - role detection of needs and business opportunities. In terms of e-mail marketing, it allows you to set up a particularly relevant contact database segmentation strategy, giving access to intelligent campaign targeting. And the ROI of e-mail marketing operations soars!

Start by evaluating the data

Today, the success of e-mail communication depends mainly on a preliminary work around the data. And advertisers have understood the importance of this issue. Indeed, the qualification of contacts is at the top of the list of objectives e-mail marketing 2013 of French advertisers (1). But how to achieve this?

Faced with multiple, scattered and heterogeneous data, the standardization and centralization of information is without question the first effort to be made. Then comes the most delicate step: identification of relevant dataand their interpretation and transformation into actionable knowledge.

This is precisely where data mining comes in, with the primary idea of making marketers' lives easier. Purchase history, nominative statistics linked to e-mail marketing operations, browsing data from one or more Internet sites, from social networks, from points of sale, etc. Data mining has this formidable capacity to rely on data from different sources without drowning in it to to highlight strategic information for marketing.

Transforming data into customer knowledge

Make the raw information "speak" and identify groups of individuals with similar behaviorsThese are the objectives sought by the algorithms used by data mining and which are so interesting. An approach that allows to push further the analysis of the data necessary to the good knowledge of the contacts and to improve the decision making of the marketers. Their choices are no longer based solely on intuition but on concrete facts!

RFM segmentation (Recency, Frequency, Amount), PMG segmentation (Small, Medium, Large customers), predictive analysis, appetence or attrition scoring are some of the most well-known analyses. In practice, what results can we expect? Identification of typical contact profilesprospects/customers with high added value (single-order, regular or VIP customers) or anticipation of future phenomena (contacts likely to unsubscribe from the e-mail program, to go to the competition, to buy this product if they buy that one...).

But beware, as for many actions, launching into data mining without a precise goal is of no interest. What are the company's business objectives? What are the difficulties encountered: attrition of the e-mail program's subscriber base, few multi-order customers, product range in decline, etc.? It is therefore clearly up to the company to define its problem. Data mining is then onlyan analytical tool that will give an answer and an operational line to follow so that it can improve, generate sales or build loyalty, for example.

Put in place real relational strategies

Today, an optimized e-mail marketing strategy includes the implementation of "intelligent" scenarios consistent with the behaviors and actions of the contacts throughout their life cycle. But considering the information already collected on its contacts as a fundamental element of its e-mail marketing strategy is a must for the implementation of coherent relational programs.

Thus, on the basis of a segmentation highlighted by data mining, more adapted to the company and in adequacy with its contacts (their behaviors, centers of interest, expectations...), the marketers have well easier to engage and develop a lasting relationship.

From a segmentation strategy comes a marketing action scenario. In e-mail marketing, this translates into the definition of targeted operations, automated and sent at the right time. This technique, called Trigger Marketing, is still not used enough in France, although its effectiveness is no longer to be proven. 64% of Internet users find this type of message more interesting to read! (2) They indeed record much higher opening and click rates than classic e-mail messages because they know perfectly well meet the relevance requirements of Internet users.

Data mining + e-mail marketing = a high ROI

Increased e-mail marketing metrics, reduced contact saturation, more effective cross-selling and up-selling, improved customer value... These are just some of the benefits of using data mining in e-mail marketing.

A concrete example to support the effectiveness of the complementarity of these tools: an e-merchant sends non-targeted e-mail messages to 97.5% of its subscribers. The remaining 2.5%, identified thanks to data mining, receive a targeted and personalized Trigger Marketing type communication. Results: these hyper-targeted messages represent only 2.5% of the volume of e-mails sent but generate 21% of the e-mailing turnover (3)!

In 2012, 41% of companies said they had a strategy to improve the management and use of their data (4). Faced with this trend, data mining must and is trying to become more and more accessible. It offers today the possibility of gradually transforming direct digital communication and customer relations in order to shape them as closely as possible to the new behaviors and expectations of Internet users. But which advertisers will be able to take advantage of these incredible data sources and gain a competitive advantage before others?


(1) Dolist - 2013 e-mail marketing practices and trends in France - April 2013
(2) Marketing Sherpa's Email Marketing Benchmark Survey - 2010
(3) DMA - National Client Email report - 2013
(4) Talend - White Paper "Where is Big Data Adoption?" - May 2013

Photo credit: Harris & Ewing, Public Domain

The author

Laisser un commentaire

Your email address will not be published. Les champs obligatoires sont indiqués avec *