What is your strategy for up-selling orcross-selling to your existing clients?

When we ask insurance executives thisquestion, they often shuffle their feet and avert their eyes whilemumbling a few words that equate to, 'We could be doingbetter…'

As companies have more access to data and sophisticatedanalytics, this is actually the right answer. Never before hasthere been more opportunity for innovative methods to identify theright clients at the right time.

Understanding the problem

Before talking about how to employ amultistage offset model, it is important to understand why thetraditional approach of identifying opportunities is no longerworking.

Most of the methods being used today (such asregressive and difference models) look at data at a single point intime. It is tedious, if not impossible, to track and accountfor customer attributes that impact insurance product decisionssuch as lifestyles, values, attitudes and opinions. They alsoaren't flexible enough to add information down the road withoutcreating significant inconsistencies. This can create a lot offalse positives as marketing team plans and execute campaigns.

That means agents and brokers aren'treaching the best prospects.

In this same data, there isn't a metricthat identifies customers that may not be ready for cross-sellingor up-selling today but could be in the near future. That's becausethe current models don't have the capability to capture howexisting customers have used products over years, thus creating ahole in recognizing how to serve clients as their needs change.

The next step

This is where the multistage offsetmodel comes in. This approach works to statistically account forthe shortcomings of the traditional model used to predict whichcustomers and products at which time. In fact, it generalizes thedifference model approach in a way that:

• Is scalable to multiple time frames;
• Captures the impact of both raw and change variables;
• Can revise past predictions as new data is obtained to createincreasingly reliable estimates; and
• Accounts for uncaptured data without having to measure it.

Multistage modeling in action

Suppose you would like to predict how aset of customers will respond to a pitch for an additional ridersuch as valuable articles insurance, perhaps to cover jewelry or awine-cellar. You might look at whether a customer has insured anynew properties, the length of time the home has been insured, andwhether there is insurance on any existing high value articles suchas furs, art pieces or sports cars.

To make the prediction, we 'd run alogistic regression model at least twice — at the end of Q2 andagain at the end of Q4 of 2018 — to identify customers who aremore likely to make such purchases.

Getting the job done

If you have the people who understandthe math behind these models, then the methodology of this approachis quite simple to implement and can be used to enhance thepredictive power for any Generalized Linear Modeling (GLM) basedmodeling or tree-based algorithms.

By doing this, you'll discover thereare three groups of customers:

• Those who are ready for your pitchnow;
• Those who will be ready in the near future;and
• Those who simply are at a lowprobability of engaging in a cross-sell or up-sellopportunity.

To mine the opportunities in the middlegroup — the ones who will eventually be ready — the model canbe run periodically to create additional data that will betterpredict which customer is ready and when.

So what does this mean?

You have the capability to use the data you currently have tofind the right customer the right product at the right time. Youalso can operationalize the whole cycle to runperiodically so that it is data-driven and delivers maximum returnon investment.

That, after all, is what we're all striving for.

Dheeraj Pandey is assistant vice president at EXL Analytics,a provider of data analytics solutions to financial organizationsincluding P&C Insurance firms. To reach this contributor, sendemail to [email protected].

This article is based on anupcoming webinar, “Application of multistage offset models incross-selling and up-selling Insurance products” scheduledfor May 3, 2018 at 12.30 p.m. EST. Here's thelink to register and attend: http://bit.ly/2HS9viJ

See also:

Here's how hybrid service models give insurers anupper hand

From model ready to business ready: Connecting datato ROI