Sales occur when consumer demand meets supply. In today’s competitive marketplace every organization is looking for the opportunity to spur further demand and many are quickly realizing that there isn’t a short-term or quick fix. It’s not enough to have the right product; marketers need to craft the right message, at the right time, and deliver it the right way to a targeted audience. By applying predictive analytics to influence consumer response to a sales offer, organizations can better understand a consumer’s preference for how the message is crafted, inclination towards a specific value proposition, and threshold for frequency of communications.
The process of building a predictive model is analogous to speed dating. Participants often have limited information on the person that the dating service has provided – perhaps just biographic background or details on common likes or dislikes. Once the “dating” starts – each person presents his or her value proposition to a certain number of prospects to see who responds, essentially testing the waters. Based on those interactions, each individual then comes up with a refined list of prospects from which they will further narrow the field into the most promising candidates.
The same approach can be used when applying predictive modeling to a sales situation. Organizations typically buy data on a list of prospects and then narrow that list to a subset which maps more directly to their value proposition. That refined list serves as the foundation for a test audience, giving marketers the opportunity to experiment and measure response in order to determine what truly resonates. This process helps organizations gather response data from which they can build a predictive model or target profile to go after.
Organizations are often saddled with the same quandary with regard to their predictive modeling approach, “Should I build a model to target people who look like the people who have bought the product or offering before?” The answer varies. If your business is mature and the brand is well known, fairly broad brand advertising may do the job. Building a look-alike model will essentially attract consumers who would have bought from your organization anyway. This saves the expense of additional marketing dollars. On the other hand, if your organization is relatively new and hasn’t yet had an opportunity to drive brand awareness, then a look-alike model may actually give a meaningful lift in consumer response.
Another common question is, “How frequently should I communicate with my consumers?” The short answer isyou should communicate as frequently as you see a marginal lift in consumer response. Admittedly, this can be complicated to nail down as it will be very volatile across time. For example, the consumer who just had a baby will not respond to your mail even if you send it twice a week (per the consumer’s preference) simply because he is busy with a new baby. Once the baby is six months old however, he may now finally have a moment to look at your mail. In this case the consumer’s demand has not changed, but the circumstances and his threshold for communications have.
There’s also the competition to consider. Perhaps it is determined that once every two months is the preferred communication timeline. How often might competitors be targeting the same consumers? Are the communications happening during the same two months or on alternating months? These scenarios underscore the need for constant testing to figure out how developments in your target consumer segment impact your outreach strategy.
Predictive analytics can give an organization the means to stay abreast of the ever-changing needs of consumers. Just like the speed dating scenario, testing and refining your specific value proposition in the market can produce a more successful outcome enabling your organization to avoid sales fatigue, grow your consumer base, and create stronger, more profitable customer relationships.View all Blog Posts