Making Segmentation Work
March 12th, 2007
Six Questions with Westfield Insurance
CMB: Thanks for taking some time to talk with us today. Can you tell me a little bit about Westfield Insurance and where you were hoping the segmentation research would take the organization?
Lisa Kozlowski: Westfield Insurance is a property and casualty insurance company in Ohio. We have been in business since 1848, writing personal and commercial insurance through our network of independent agents. Up until three years ago Westfield didnt have a formal marketing research department and had never done any type of true market research. With so many changes in the personal lines market in terms of competition from direct writers like Progressive, etc. who were starting to write a lot of businesswhereas our customers cant call a 1-800 number or go online and get a policywe needed to look at the future in the personal lines arena and figure out what the potential was for Westfield.
Susanne Wahl: Well never be going after the price sensitive customer. We also dont plan on changing our distribution channel. Given those two key things, if we are going to continue to stay in the personal lines market, we needed to find out how we could maximize share of wallet, but also do that within the framework of who we are and who want to have as customers.
CMB: So with your segmentation, were you hoping to find new segments specifically or did you just not know how your target marketor potential targetmarket really looked?
Susanne: Due to legacy systems and the fact that our data is at the policy level (not at the customer level), we didnt really have a good, defined picture of who our current customers were. Therefore, we were not sure who to target, how to appeal to them, reach them, etc. We had quite a few gaps in our knowledge base. So, we needed to know who was out there, what they liked, and if we could realistically reach them given our capabilities and resources.
CMB: Realistically seems to be the key word. What steps did you take to ensure that this research would produce something you could easily use?
Susanne: Several things were key to making sure the research was useful.
1) We got involvement from executive leadership right away.
2) We asked them exactly what decisions they were planning to make with this information, which ones were most important, and if they were prepared to act on this information when it came backeven if it came back in a negative way.
3) Working with Chadwick Martin Bailey intimately during the analysis portion ensured that all of the potential solutions were discussed and that they accounted for our need to stay within the independent agency channel. It also ensured that we used a solution that optimally fit the data as well as our value proposition, our mission, and goals.
Lisa: It really was critical that we had support from the top levels. We knew that this was going to be the foundation of personal lines strategy moving forward. We knew it was going to be more than just a marketing exercise.
Susanne: We let them know that this was not going to be a black and white solution with a perfect roadmap and that they really were going to have more work to do once the research was completed. Personal lines leadership was going to have to evaluate what we got back and determine from a product, underwriting, marketing, etc. perspective which segments they were going to go after.
We had to step out of the role of research to encourage and assist people to use the information: What do they do next? How to build a plan with it? So we went into a planning role which was challenging, but it was worthwhile to make sure that we did provide value with the information we got back.
CMB: What can your end users do now that they couldnt do before?
Lisa: First, we are able to have more confidence in building an actual strategy that is customer focused. Weve always had a goal to be customer focused in previous years but we never had strategies and plans to support that.
Second, we are able to understand more about how our current customers work and feel confident that there is market potential to sustain us for some time. The biggest benefit was highlighting the opportunity to cross sell and maximize the return we could get from each customer.
Also, instead of just assuming that a customer still wanted to buy through an independent agent there was validation that there are still people who connect with our value proposition, which includes selling through an independent agent.
Susanne: It also allowed us to learn more about customers than we already knew in a sense of their lifestyles, what their motivations are for finding insurance. For the first time we could see where the potential is so we can align our distribution appropriately for maximum effectiveness. Now when we talk about what the key messages are, we have a better idea of how to position ourselves to appeal to the right people.
CMB: How would you say that doing segmentation in the insurance industry is different from a more product oriented market or just other industries?
Lisa: I think that there are three areas that insurance is different.
First, there are more things we need to consider when looking at the actionability of a segment, for example in insurance you not only have to evaluate the potential of someones wallet, you also have to look at and evaluate risk factors. So the people with the most money and the most potential are not always the best insurance risks. The segmentation might look great, but then we could look at additional information and they may not be the type of people we want to underwrite from a risk perspective.
Another thing is that insurance is often seen as a commodity product and many people are not too involved in their decision. So that is why incorporating our distribution channel into the design of the segmentation was key.
The third thing is that in the insurance industry redlining is a big thing. We have to be very careful and politically correct, we have to be very sensitive to whom we target in terms of race, neighborhood, and things like that, and those things really have to be more&
CMB: Do you have to take that out of the equation?
Susanne: We have to in terms of defining the segments. However, we can use things to describe each segment. We have to be prepared to show proof that we are not targeting based on anyones race, neighborhood, etc.
Lisa: We have this established distribution system. And, over the years, we have relied on those agents to bring that business in, so now there is an entire shift for the independent agent system and regional carriers like ourselves to start defining our appetite. Rather than saying, Agents, we want to write all the homeowners we can, we want to say things like, We want to write customers that are align with our value proposition.
Susanne: I think too that the underwriting and risk aspect requires a lot more involvement from a cross-functional team than what you would do in other industries. For example, most companies typically use the market segmentation mainly for creative messaging and reaching folks with those messages. We were using it to drive long-term strategies, not just tactics. This requires involvement from many different areas of operations in addition to marketing.
CMB: Is there any advice you would offer to anyone in a similar situation or anything else you would like to add?
Lisa: Direction needs to come from and be internalized at a corporate level – a senior executive, president, and CEO level. I think defining very clearly what to expect from this process, what will and wont come out of it is so critical. You dont want to be left at the end with people envisioning that you had this project and you would find leads and customers and they were all so perfectly aligned in these wonderful neighborhoods. You have to be real clear with what you are communicating, I think you need cross functional involvement – there is absolutely no way you can do the research, hand it off to an executive then walk away.
Lisa: Most importantly, you have to partner with a segmentation expert, not just a research company who happens to do segmentation. You have to find a company that specializes in segmentation.
Susanne: They need to customize their approach and not just offer a canned approach. I think CMB brought a lot of value to the table in terms of analytical skills, but really, if they did not have a familiarity or a theoretical background about how to apply different statistical methods like CHAID and discriminant analysis, factor analysis, they would not have known which solution provided the best fit for us.
Lisa: Or to design the survey properly to get there.
Susanne: That was really key because a lot of times people try to sell you a canned approach and they are showing me that they are more concerned about selling their product and not necessarily concerned with meeting our organizations needs.