Matt White, writing on Finextra, shares Turkish bank Garenti’s CRM strategy:
If you go into their branches you have to swipe your bank card receive a ticket, and wait to be called. But not all tickets are equal. High income ‘superstars’ get seen within five minutes, the pretty well off have to wait a bit longer, and the great unwashed should be granted an audience after about ten minutes. Non customers have to wait for up to half an hour. Garenti says that if these people want to be seen sooner, they had better open an account.”
My take: Funny, but instructive. Funny cuz’…well, you know why. Instructive, though, because it highlights two issues that banks, often unknowingly, have:
1) Their customer segmentation scheme does more harm than good. Although few (if any) North American banks require customers to “swipe in” before receiving service, the Garenti story does highlight a drawback that many banks here do face. Few banks, thanks to how they segment their customer base, account for a customer’s potential relationship value.
Like Garenti, many banks segment customers on number of products owned and demographic factors like income. But number of products currently owned may not provide any insight into how many more products a customer might have with the bank. And a customer’s income, while a potentially good predictor of the need for financial products, isn’t necessarily a good indicator of the products that customer will consider the bank for.
As a result, this approach to segmentation may lead to preferential treatment for customers who have no intention of expanding their relationship with the bank, and reduced service levels to (and attention to) the customers who represent their best long term opportunities.
While many observers complain that segmentation schemes are, often, not actionable, there’s another issue. Most approaches don’t help the bank understand the kinds of relationship a customer wants to have. Some customers are looking for a personal relationship with a banker, another wants help and guidance making smart financial decisions, while a third may only want to park his paycheck into a checking account before putting that money in investment accounts (that aren’t with the bank).
In Garenti’s case, it’s possible that some of its “best” customers might tolerate a longer wait in the bank, especially if they could come in, get a cup of coffee, and relax for a bit. Isn’t this what a lot of banks are going for, trying to re-create the Starbucks experience in their branches? But Garenti’s segmentation dictates a CRM strategy that forces it to ignore this option.
2) They fail to understand the importance of the initial sales experience. Garenti’s tiered service levels may improve service to its “best” customers, and possibly even improve its retention levels. But it may adversely impact new sales (i.e., prospects getting up and leaving during their 30 minute wait), and — just as importantly — be hurting future customer loyalty. The longer-term loyalty impact is not as apparent because few financial firms really understand how important that initial sales interaction is for shaping future purchase intention.
Opening a checking account may not be a particularly stressful, or highly emotional, event for many customers. But it’s likely that they’ve made a conscious choice to pick one provider over another — and there is stress or emotion involved in worrying about whether or not they’ve picked the right one.
While a bad experience in that initial interaction may not dissuade a customer from doing business with the bank, it might produce a negative story that the customer tells himself about the bank. A story that could limit the future potential of that relationship right from the start by diminishing the customer’s desire to turn to the bank for more products and services.
So should Garenti — or any bank for that matter — flip-flop the scenario and give prospects priority? Well no, that won’t work either. Then they’d be no better than the telcos who give all the good deals to new customers and treat existing customers like 2nd class citizens.
_________________________
There’s no easy answer to these issues. But, as a starting point, banks must: 1) better understand the kinds of relationships that their customers want to have, and 2) develop a segmentation approach that builds on those relationship types. And not simply let the existing segmentation dictate their CRM strategy.
Technorati Tags: Marketing, Banking, CRM, Strategy, Garenti, Finextra
Ron Shevlin's Marketing Whims 
I thought I was seeing things when I read that potential new customers had to wait the longest! My heart goes out to the marketing folks who did their job and brought these new prospects into the bank to open an account. Only to see them leave 15 minutes later when they get a glimpse of how nicely their new friendly bank treats them. Of course, I shouldn’t be too sad for the marketing team–after all, they probably also designed the segmentation scheme driving this treatment…
I agree that there is not one easy answer. Oh, if only we had that crystal ball to tell us exactly the type of relationship any given consumer wants. The icing on the cake would be an absolutely accurate prediction of lifetime value and long-term loyalty.
Overall, it is important to constantly evaluate your segmentation scheme to make sure that it is addressing today’s market and today’s goals.
Just because a particular segmentation scheme is bad does not necessarily mean that segmentation schemes IN GENERAL do more harm than good. A good segmentation scheme should reflect the bank’s strategy so, for instance, a bank focused on cross-sell should have a segmentation based on propensity to buy additional products. Similarly, a bank focused on retention would segment based on current business. Applying these segmentation models, taking action based on them, requires thought and can do more harm than good if done poorly. The need to trade off various predictors and other data items is what makes this complicated, of course!
I would also say that some kind of adaptive control ( http://www.edmblog.com/weblog/2007/04/why_do_you_need.html ) is essential – some way to try different approaches and see what the difference is. For instance, perhaps a subset of customers would be prioritized based on a different approach and the impact of this tracked.
These charts from McKinsey quantify what the impact of those “negative stories that the customer tells himself about the bank” might translate into – very instructive !
http://www.mckinseyquarterly.com/newsletters/chartfocus/2007_05.htm
The reason segmentation is not actionable is because organizations develop segmentation schemas in silos. Different groups develop their own segmentation scheme and the same customer gets multiple segment scores based on whether direct marketing is trying to target him or customer service is figuring out their gold,silver and bronze customers. What is more – almost all the time – none of the segmentation schemas are linked to market research – so we learn about the customers but do not map the learnings back to customer segments. This is the fundamental problem – which we describe as the elephant and blind man syndrome in segmentation. A more robust platform approach is required by marketing groups who need to bring together various factions of marketing under one segmentation umbrella using some standardized schema. We wrote a whitepaper about the topic recently, which can be downloaded from the blog.
http://diamondinfoanalytics.com/blog1/2007/05/24/rapid-market-segmentation/
Jeremy: Thanks for the link. Those are great data points.
Amaresh: Great points, as always. Your points really support why I haven’t been a fan of “persona development” with Web site development. It’s just one more segmentation approach, disjointed and detached from other marketing and segmentation efforts.
Pingback: Will Work for Data » Marketing Productivity Blog » Blog Archive
Interesting.
I would say the way segmentation is implemented in customer service is wrong and you are correct in pointing that. Segmentation has to be used in target marketing on campaigns i.e. to focus on a segment more than others. This does not apply on the service to be given to people who come to the bank for any work. All the customers need equal treatment and that is the basis for long term good will. Any unfair treatment will lead to loss either in revenue or in brand value.
Bhupendra
Interesting you mention this Ron.
I just read an interview in India with a seasoned CEO of a leading Japanese automobile manufacturer in India. He had mentioned that they had implemented a “high value customer, Service now” strategy.He had explained to his dealer that this was a great idea.Seperate product desks to ensure quick service etc. Hi value products need exclusive service, it seems.
In reality, there were customers waiting in queues for other products and this desk, I believe was empty. When customers asked at the counter, why the service rep can’t service the request, it was mentioned it was only for select customers! This pissed off other customers leading to questions in their minds that am I a valued customer for them at all!!
It was removed instantly. The dealer’s premise was in every customer’s mind he/she is valuable.We need to find ways of serving each of them rather than making it so explicit!!
Interesting views!
When you start treating all the customers alike, are you not ignoring the best customers in a way? Is it not logical that you need to invest more on best customers hence better ROI? What would you accomplish if you spread your marketing dollar homogeneously among all the categories of customers that include rogue and non-profitable customers? Wouldn’t that hurt your ROI?
This in turn would weaken the brand value’s link with shareholder value, thereby eroding both.