How to Use Predictive Analysis for Target Marketing - dummies

How to Use Predictive Analysis for Target Marketing

By Anasse Bari, Mohamed Chaouchi, Tommy Jung

Predictive analytics make your marketing campaigns more customer-oriented. The idea is to customize your advertisements to target a segment of your total customer base — not the whole. If you send only the ads that are relevant to a segment of customers, you increase the likelihood that those particular visitors will perform the action that you hope for — buying.

If you can determine which segment of your customer base will respond best to your message, you save money on the cost of convincing a customer to make the purchase (acquisition costs) and improve overall efficiency.

For example, when you pay an online ad network — for example, Google AdWords — to display your ads, typically you pay for each click that sends traffic to your website through a sponsored ad that appears in response to a search.

Getting the visitor ultimately to do what you hope she’ll do while she’s on your website — become a paying customer — should be part of your marketing strategy. This type of marketing cost structure is called pay per click. You pay the network (in this case, Google) for each click, whether or not the visitor converts into a sale.

Because you are paying for each click with no guarantee of converting each visit into a sale, you’ll want to create some sort of filter to ensure that those likeliest to become customers receive your advertisement.

No point displaying your ad to just anyone — a shotgun strategy is far from optimal, and your acquisition costs would be through the roof. Your ad’s target audience should be those visitors who have the highest chance of conversion.

This is where predictive analytics can come to your aid for target marketing. By creating an effective predictive model that ranks the customers in your database according to who is most likely to buy, subscribe, or meet some other organizational goal, you have the potential to increase the return on your marketing investment. Specifically, predictive analytics for marketing can

  • Increase profitability

  • Increase your conversion ratio

  • Increase customer satisfaction by reducing unwanted contact

  • Increase operational efficiencies

  • Learn what works (or doesn’t) in each marketing campaign

Traditional marketing targets a group of customers without applying such modern techniques as predictive modeling using data-mining, and machine-learning algorithms to the dataset. Predictive modeling, in the area of direct marketing is called response modeling using predictive analytics (or simply response modeling from here on). Sometimes analysts create filters to apply to the dataset, thereby creating a select group to target.

But that select group may not be optimally configured. Response modeling, on the other hand, seeks to discover patterns in the data that are present but not immediately apparent; the result is an optimized group to target.

The following example uses a small sample to compare the profit generated by direct mailings — traditional marketing versus response modeling.

Traditional Marketing Response Modeling
Number of customers targeted 1000 100
Cost per customer targeted (assume $2) $2 $2
Number of responses 20 10
Response rate 2 percent 10 percent
Total revenue (assume $100 per response) $2,000 $1,000
Total cost of campaign $2,000 $200
Total profit $0 $800

The response modeling has targeted 10 percent of the traditional number of customers (100 instead of 1000) to an optimized subset. The response rate should be higher with response modeling — 10 percent instead of the 2 percent that is typical for traditional marketing.

The net result is a profit of $800 under response modeling; traditional marketing breaks even. Also, as per-customer targeting costs increase, response modeling’s value gets even better — without even taking into account the implicit benefits of not targeting unqualified customers.

If you make constant contact with a customer without providing any benefit, you run the risk of being ignored in the future.

Now let’s consider an example that shows the profit comparison between direct mailings using traditional marketing and response modeling with a larger sample size.

Traditional Marketing Response Modeling
Number of customers targeted 10000 1000
Cost per customer targeted $2 $2
Number of responses 200 100
Response rate 2 percent 10 percent
Total revenue (assume $100 per response) $20,000 $20,000
Total cost of campaign $20,000 $2000
Total profit $0 $18,000

Here the response modeling has (again) targeted only 10 percent of the 10,000 prospective customers traditionally targeted. In an optimized subset of 1,000, the response rate should be higher. If you assumed a response rate of 2 percent for a traditional direct-mailing marketing campaign; with response modeling, the response rate is 10 percent because the customers are likelier to buy in the first place.

Response modeling creates a profit of $18,000 under this scenario; traditional marketing breaks even. As in the previous scenario, any revenue earned using traditional marketing is consumed by marketing costs. Thus, as the accuracy of customers targeted, increases, the value of response modeling also increases.