Performance Evaluation Case Analysis

Citibank

Financial -- Above par

Strategy Implementation -- Above Par

Customer Satisfaction -- Below Par

Control -- Par

People -- Above Par

Standards -- Above Par

Overall Evaluation -- Above Par

The overall approach for feedback is to be honest. You are an adult, dealing with an adult. We need to treat it like a management issue, not a human resources issue. Optimizing the customer service score, for example, may not be the best way to run the branch.

The approach is to walk through each of the points on the form one by one initially, giving more or less equal time to them all. So there will be an initial, structured sequence that we use, and we would want to use the same one for all of our managers. But after that, the discussion needs to be unstructured, because that is where the really good exchange of ideas comes from.

Then, because customer service is the only Below Par rating, that will receive special attention. First, James probably understands the point of improving customer service scores, but it is worth reviewing this. Customer service is linked not only to corporate priorities but it serves a specific purpose within Citibank's strategy. The branch performs well financially because James is a good manager and because there are a lot of wealthy people in the client base. It is those wealthy individuals that drive the branch's financial results. But competition for those customers is going to be intense. Strategically, the key to sustainable success at this branch is to sustain high customer service scores, especially among the wealthiest clients.

Here is the rub -- James is a pro, so he is going to want to understand the data a little deeper. First, there was a lot of variability in the customer service scores. The "Below Par" rating is because, on average, the score was below par, but there is a fair bit of volatility in the scoring and we need to recognize that. If James is going to improve, we need to be able to help him with meaningful information, not just a letter grade.

The bottom line is that the methodology is weak -- it is not aligned with the overall strategy for this branch. This is why his overall evaluation is going to be "Above Par." A survey of 25 randomly-selected customers makes little sense for a branch where the top 10% or so of customers is driving revenue and profit. Further, they are probably the busiest customers, and the least likely to answer a survey. The customer service scores -- in the absence of actually having this data -- may well reflect that the branch is driving its financials by delivering exceptional service to the high net worth customers, perhaps at the expense of the low net worth customers. If that is the trade-off that James is making, we need to know that, so this will have to be part of the discussion. While all branches have x percentage of clients driving revenue, this branch is likely to have a disproportionate amount. I would measure this branch against other branches with equivalent customer bases -- so ritzy residential areas, or financial district branches in New York, Chicago, San Francisco, etc. You really have to compare like to like.

The desired end result of the discussion is to understand why the data is the way it is, and to get a better sense of how James is approaching this issue. Naturally, we want even the little old ladies to be happy with our branch, but we need to be aware of the trade-offs and ensure that what James is doing in the way of trade-offs is not negatively affecting the shareholders' wealth. If he needs an extra ATM or an extra teller to make all the little old ladies and clock punchers happy, that's our job as regional managers to make that happen -- the branch is making a lot of money for us so it is not a problem to provide them with the resources that they need. But this is the sort of thing that will come out of our discussion with James about the customer service issue.

3. The discussion of the issues at James' branch has revealed that there are some problems with the feedback system with respect to customer service at our branches. First, 25 customers per branch may or may not be enough. At diverse branches, this might not be sufficient to be a representative sample. For example, at branches where there are a lot of high net worth individuals, you must ensure that some of them are included in the survey. These people tend to be busy, thus less likely to respond to a phone survey, and that is assuming you could get past their assistant to even ask them if they want to do a survey. So chances are pretty good that a 25-person telephone survey is not getting a representative sample. For James' branch, and others with a similar customer base and revenue drivers, this matters.

So sample size might need to be bigger, but Citi will also need to find a way to reach the high net worth clients for customer feedback. At branches like this, how they are serviced is far more important to the bottom line than how the average client is serviced, because they account for such a disproportionate amount of revenue and profit. Finding ways to include them in the feedback is critical as far as getting feedback that matters is concerned. So the sampling method will need to be different.

Lastly, I would exclude factors that are out of the branch's control. This was a bigger issue with some of the other aspects of the performance evaluation -- the Control section in particular -- but the survey was based in part on things that are usually out of control of the branch manager -- ATM s (maintenance is usually a centralized function, and contracted to a third party anyway) and the 24-hour phone line (which is definitely centralized). The branch should not be evaluated on things that the branch does not control. Such feedback is useless because there is nothing that the manager can do to remedy the situation.

Another issue I would raise is the issue of feedback timing. In this scenario, James has received feedback quarterly on his branch's performance. He has already heard about the poor customer service scores and taken action. Given that, what value is there in further punishing him for those results? It is logical that you cannot use one quarter's results but at the same time, the quarterly feedback has already been accounted for in this process. The time and energy that goes into performance evaluations would be better spent if the feedback was current, and relevant to a discussion today about where the branch can go in the future.