Financial Client Satisfaction Index
Sampling Your Customers This page of the web site is important. You may want to print it out and discuss it with others, so that you agree on your preferred methodology.
The most important aspect of selecting a sampling methodology, is to use it consistently. Give careful consideration to the method you use to survey your clients. You must stick with this method month after month so that your trend lines are comparable.
Whichever the method you choose, there is going to be some bias. That is, the results will reflect certain types of customers contingent on the method you choose. For example, use of the Lobby method will produce a different bias than use of the On Line method, and the On Line method will produce a different bias as opposed to the the Direct Mail or Stuffer method. What's important, however, is consistency. Pick the method (or combination) and stick with it.
We would recommend that you use a combination of the On Line Survey, the Lobby Survey, and either Stuffer or Direct Mail.
If you have questions or need assistance in determining your methodology, please notify us us and request that we contact you by phone.
We suggest you appoint an FCSI Contact and Coordinator to oversee the project and serve as a liaison between your branch and our service representatives.
Lobby Method With this method you place the lobby questionnaire in a prominent, visible location; or you have your staff (probably the branch manager) pass them out at different times during the month. Or, most probably, both. For example, it might be a good idea to have the questionnaires available at all times, but also have the branch manager pass them out at more active times of the month such as direct deposit day, government employee pay days, Fridays, or specified "slow" times. Whatever schedule you choose, be consistent!
The collection box should be highly visible. Perhaps on the check writing counter or teller counter ledge. Once you select a good location, keep it there.
It would be a good practice to have your branch managers retrieve the completed questionnaires either daily or weekly and look at the comments section. Have them make notes on anything pertinent which can be addressed in the branch and take quick action (e.g. the ATM is out of deposit envelopes or is dirty). These comments can also be good for weekly sales meeting discussions and for real time remedies. We have prepared a simple Log of Customer Comments for you to print and use.
You should establish a "cut-off day" for your month which will allow you enough time to tabulate the questionnaires (see processing) and enter your data into the database. For example, you may decide that you want to have a sample of 300 questionnaires each month to enter into the database to get your reports. And that it will take you (or a clerk) two days to enter the data from the questionnaires into the upload template. If this is the case, you may decide to make your cut-off the third Friday of the month. This will give you a week to do the upload work! This does not mean the collection process stops; just that you ask all branch managers to send all questionnaires in by the third Friday of the month.
You should shoot for about 300 questionnaires to make up your total sample. If you decide you want reports at the branch level (or other segments), you should shoot for about 25 to 50 per branch. This, of course, will probably boost the size of your total sample, but that is OK.
It is probable that at the end of the cut off period you will have many more questionnaires that you need. That's OK. You can "sample" these to get back to your target number. For example, if you get in 1427 questionnaires on the third Friday, but you only want 300, you sample the 1427. To do this establish your sample interval, by dividing the desired number in the sample (300) into the total you have (1427). This produces your sample interval of 4.8. Round this down to 4, and choose every 4th questionnaire from the deck of 1427. Be sure to shuffle these before choosing.
Statement Stuffer or Direct Mail These methods have the advantage of reaching those customers who may not visit your branches as much. However, it is more costly, in that you will have to pay for the return mail. However, if you do your sampling well, you can keep this cost way down. If , for example, you have 20,000 checking accounts, you may not want to stuff all of them or send all of them a letter each time you do the survey (monthly or quarterly). Rather, stuff or send only enough (using the sampling interval approach described above) to get back the three hundred or so you want. Experience suggests that statements stuffers will get a significantly lower response rate that targeted mail. This may vary , and you should experiment, but a starting point might be:
10% response rate for statement stuffers
25% response rate for direct mail.
Using these figures, if you have, for example, 20000 checking customers, and you want 300 returned surveys, you should stuff 3000 statements (300 /10%). This would produce a sampling interval of 6.7 (20000/3000), which you'd round to 6 and stuff every 6th statement. In the following months, you'd do the same, but start with a different customer. This would insure that the same customer was not sampled every month.
You would use the same approach to sampling your customers for direct mail. But since the response rate is higher, you would need to sent out fewer (300/25%=1200).
Telephone Survey You may prefer to use a telephone survey. While it is potentially more expensive because you will have to hire professional interviewers, there are offsetting advantages also. You can more precisely control the size of your sample and its "randomness". You can also phone non customers and gauge the satisfaction they have with their banks. The language flow and pacing of a telephone survey is critical and we recommend that you not try it yourself, unless you have the experience.
If you would like to consider a telephone survey, e-mail us and we will contact you to discuss this option. We can also advice you on a professional survey company to do this for you. If you are already doing periodic telephone surveys, you might wish to consider incorporating our questions, so you can compare your satisfaction scores to the benchmarks.
Sub Group Sampling (Branches and Special Segments) This refers to a "sample within a sample". For example, you might want to see how the customers of different branches view the bank. Or you may want to see how different types of customers (MCIF groups, such as deposit deciles) view the bank. And then compare these segments or sub groups with the total sample.
Our database is structured so you can input data by branch and/or segment. You can then see these results separately (via "pull down" menus) and compare them to your overall base.
In the case of branches, customers are asked to put in the name of the branch when they fill out the questionnaire. In the case of special "used defined" (e.g. deposit deciles) segments, you will have to pre code your survey forms before you send them out, because customers won't know this. There is a little box for this on the survey form itself.
What's important here, is that you don't create a bias by having too many questionnaires from one branch (or segment) versus another. With branches, this is fairly easily controlled. You can simply make sure that when you upload, you have approximately the same number for each branch (e.g. 25 to 50).
If you plan to use direct mail ( which will be necessary to implement MCIF sampling), you also need to make sure that you don't overbalance one of your segments. The rule here is simplicity. Don't use too many segments. It's smarter to just not include segments which you know to be unattractive, than to try to manage 20 to 30 segments.
Special Note: After you sign up, you will be asked to supply names for your branches or special segments so the database will recognize them. Be sure to keep this up to date.
Employees Regardless of your approach to sampling, be sure to involve your employees in the FCSI project from the beginning, so that they are positive with customers. Explain that this is an ongoing program to provide feedback to management. And that the result will be improvements in customer service and satisfaction.
Also, be sure to share the results, especially the trends, with employees to show teamwork and progress.
It is probably not a good idea to use FCSI results as a formal component in branch employee performance evaluations, especially if the cards are returned to branch employees. Mystery Shopper programs and other metrics are better for this. However, as a manager, you can certainly gain some valuable insights as to which of your branch staffs are doing better than others; and the results can be useful in your informal coaching.
Customer Comments This section of the questionnaire is very important. And while the comments are not trendable or quantifiable, that doesn't mean they should be ignored. Branch managers should note them down each week and discuss them up and down the reporting chain. Supervisors should see them and/or ask about them. The should be brought up in executive management sessions.
We have created a comments log sheet to help in this process. Make sure this is distributed and used. The best way is to get the CEO involved. Help him/her by make sure its passed along and made the subject of discussion. This may be difficult, because "ox's will be gored", but it can be very productive for customers.
Recommendation Based on this discussion, here are our recommendations for starting your FCSI effort:
1. Keep it simple. The purpose of the survey is to measure and track satisfaction over time. We suggest that if you use sub segments, you limit them. If you have under, say, 100 branches, you can probably look at them individually and compare them to your overall average. If you have more, you might want to put them into "markets" (e.g. rural vs. urban, or cities, or responsibility centers). If you want to look at MCIF segments, look at no more than 4 or 5. Remember, when you sign up, you will be asked how you want these defined! Think this through and stick with it. E-mail us if you need help.
2. Be consistent. Once you set up your sampling and data collection method, don't "fiddle" with them. The trends are the key!
3. Don't use FCSI as a direct performance measurement component. If you want to reward branch employees for performance, measure such things as: customer retention, sales, cross sales, etc. All of these direct measures will be facilitated if your customer satisfaction is improving. Once you have these direct measures in place, then use FCSI results as a coaching tool to help branch people achieve their performance goals. They will do it naturally, anyway.
4. The lobby method is good. Even if you decide to do mailings or stuffers, we recommend that you deploy the questionnaires in the lobby; and have your managers pass them out on a disciplined schedule. The PR value is immense. And by putting your managers in the "breach", they will learn first hand what is working and what needs repair .....at a very local level! Ask to see their client comments log on a weekly basis!
5. Make sure your CEO participates. Go over the results each month, just like the financial. Set goals, develop strategies and plans, follow up. Share the results and actions with employees. Make improving customer satisfaction a crusade! Remember, this is a journey, not a destination!
next