The Truth About Customer Satisfaction

The Truth About Customer Satisfaction

We started measuring customer satisfaction on emailed support tickets a little more than a year ago. As an enterprise software company, customer support is a large part of what we do day-in and day-out. We realize (and embrace) the impact it has on the perceived value we create in our products and services. Because of that, we wanted to be able to measure customer satisfaction in near real-time with a system that worked with the customer’s preferred communication channel.

The System

The system works like this: An email comes in from a client, at which point a support ticket is automatically generated and then assigned to an appropriate resource. After a support ticket is resolved, the client is given the opportunity to rank our service on a scale of 1-6 (1 being terrible and 6 being outstanding). If they choose to rank us (here’s the template email they receive after a resolution) they are also given the opportunity to provide additional feedback via a simple web form.

The case for implementing the new customer satisfaction system had two clear benefits:

  1. We’d be able to capture a measurable level of customer satisfaction the instant we provided a resolution to a question or issue. This would allow us to be more aware of, and immediately address, any clients who weren’t completely satisfied. It would also provide a great way to improve future service with specific examples/measurements and praise staff members who were providing outstanding service.
  2. We could eliminate some duplicate processes; namely the act of relaying a resolution to the client (typically via email) and then filling out a time sheet with a similar explanation.
  3. This would allow us to be more efficient with our time, thus providing even better service and perhaps a few more billable hours each year. More value for our clients, more revenue for our company. Win-win.

The Results

After more than a year of measuring our customer support, I gathered some results:

  • Approximately 15% of support tickets receive a ranking
  • The average ranking (on a scale of 1-6) is 5.493
  • 62.33% of the submissions were ranked a 6 (Outstanding)
  • Four submissions were ranked a score of 1 or 2 (Terrible or Pretty Bad)

There’s plenty to be learned from this type of data, but one thing stuck out to me:

Three of the four substandard rankings we received came from the same person.

That person only submitted three rankings all year, despite having 21 resolved support tickets (in line with the 15%). So what’s the obvious takeaway here?

You can’t please everyone.

And maybe that’s the truth about customer satisfaction. The question is what should be done about it? Do we focus on the complainers and try to turn them into evangelists? Some people think so, and I’m not so sure they (myself included at times) are wrong.

On the other hand, I’ve met people that don’t like Zappos. I know people that can’t stand Apple. I’ve talked to one or two that refuse to fly Southwest. So where does that leave us?

Maybe the answer is to spend more time learning from the customers that love you and less time trying to please the ones that don’t. What do you think?

Enjoy this? Get the Newsletter.


18 Comments


  1. I totally agree with your assessment: there are just some people you’ll never be able to satisfy no matter how hard you try. You’ll find much more value in nurturing and cultivating existing customers who have a secure or even vulnerable relationship with your company than the ones considered “at risk”.

    However, you need to do the due diligence to understand who these “at risk” customers are. Are they your most tech-savvy customers? Do they spend a lot of money with you? Are they influential in the social media community? These things can have an impact on steps you decide to take.

    Influential customers who are your biggest critics often times are looking for some extra “love” from your company. For example, we have reached out to some of these customers at times with offers to have them talk directly with our product managers to voice their concerns about our product. Not all of them take us up on the offer, but the ones that do are grateful for the experience and often times become active (and vocal!) supporters.

    It’s always better to spend your time and effort “upstream” with your customers building a secure, value-added relationship. But pay attention to those “squeaky wheels” as often times their feedback can be gold!

    • Thanks for the comment Larry. We take feedback very seriously; it was one of the reasons we implemented the system in the first place.

      You’re right about squeaky wheels, but it’s also nice to know which ones will always squeak, regardless of how much oil you throw at them.

  2. Great post, Tim. I agree that you can’t please everyone, and you’ll go broke trying. We’ve found that if you focus on improving the things that are issues for your promoters, delighters, fans, or whatever you want to call the group that loves you, you’ll get a lift from everyone else. Having said that, there will be that group of people that will never be satisfied, no matter what you do. The time will come when you’ll need to decide if they are worth the effort or if it’s time to fire them. Weighing into any and all decisions, too, should be the value of that customer, i.e., the 80/20 rule.

  3. I’m wondering whether the “real takeaways” aren’t being taken away here.

    First, your data means nothing and can’t be interpreted. Period. 15% of tickets rated? Biased sample. What you SHOULD take away from this is that for whatever reason, customers had limited motivation to bother rating. THAT might be the real issue here for your business.

    Second, surveys and rankings are misleading at the best of times. Did you have a survey professional make up the questions? Were they tested properly? WOuld the results be both valid and reliable? No? Of course not.

    Third, the “average ranking” means nothing in business terms. It’s, in general a poor indicator of anything for this kind of data. It’s the distribution of data points that has business significance.

    Fourth, doing a “mean” on ratings is not acceptable practice for this type of data. Median score, ok. not Mean.

    These problems are typical in the customer service arena. It’s just plain sad.

    • Charming as always Robert.

      I agree, 15% is lower than I initially expected. It’s obvious that people’s motivation to respond is going to be higher when the service is really great or really bad. I’m not sure how to get around that though…I would be open to suggestions.

      No, we didn’t have a survey professional make up the question. Not sure if you read the whole post or not, but it’s just one question. I don’t think we need a “pro” to come up with one question.

      The median was 6…not sure that’s any more valid than the mean.

  4. Good post, and good comments thus far. I will add one simple point, but first, a little background.

    Back when we (Powerfeedback) launched our first web survey in 1989, both business in general, and market research specifically, looked much differently than today. Web surveys (and email versions) today are horribly overused, and in more cases than not, very poorly constructed. Reversing an old axiom; garbage out, garbage in.

    Looking at this on the larger scale of not being able to please everybody is fine, but, also keep in mind that our firm, as well as others, have plenty of data that shows the training and quality of service providers has declined over the last 2 decades.

    As for what’s next, for the firms who “get it”, we will see more qualitative research and less quantitative. Mobile platforms such as SurveySwipe will continue to grow. 2011 will continue the sales and revenue decline of the top 25 market research firms. Sole practitioners should fare well, though not as much as the smaller “niche” or “boutique” M/R firms.

  5. At the conclusion of a recent client feedback session, the client I had just interviewed shared with me his perceptions of the value of client feedback. “Feedback is a gift. Negative feedback is an even bigger gift,” he said.

    I couldn’t agree more. While it is great to validate what one does well through feedback, the true opportunities to grow *organizationally,* long-term come from understanding and acting on the perceived negative. One may not be able to please that specific client, but the lessons learned can certainly transfer to the service approach for other clients.

    What worries me most about this case study is the 85% who didn’t respond. The adage “no news is good news” just isn’t true in client feedback. For every person who complains to you, there are at least 10 others who will quietly take their business elsewhere (and complain to others). That’s risky.

    Good discussion here! :)

    • Thanks Stephanie…I agree that the 85% is a number that deserves some focus. As I said above, I think it has more to do with being merely *satisfied* with the resolution rather than wowed.

      If someone has a simple question that’s answered in a reasonable time frame (per the expectations we’ve set), then is that going to wow them? Probably not. If someone has a complex problem they’ve been trying to figure out for 2 weeks and we can solve it in a day, then they’re much more likely to give us a rating.

      I just don’t know how we can have any control over the respondents’ desire to click a button in an email. As I said to Robert above, I’m open to suggestions.

      • Prizes, bribes from your dad (below)? Hehehe! Seriously, though. I’m not sure what would work for your organization. I’ve seen in the past departmental representatives advocate for the survey in various scenarios, whenever it made sense. As part of an elevator speech, as part of a presentation to management, an article in a newsletter, etc. Some folks just don’t understand how responding can actually eventually benefit them. Good luck!

  6. Feedback is a pretty important tool in the business, especially if you’re aiming to grow from your mistakes and improve great achievements. You’re quite right though, although you may not please everybody, it’s important to listen and perhaps we can learn from their complaints.

  7. Good stuff! Of course, I’m somewhat bias since I’m your dad.

    I definitely will pay more for good service and will seek out those who provide it.

  8. Hey Tim,
    Nice Reply to Mr Bacal. You’re a class act, my friend. I wonder if your customers, especially the ones on the fringe, high and low, gave you additional information that would be actionable for you. What particularly was exceptional or poor. Also, what was your follow up plan to elicit more detailed information that would help you address the root causes of the successes or failures?

    Thats always the challenge.

    Cheers
    Barry

    • Hey Barry, thanks for the comment.

      Yes, customers that respond to the one-click survey are directed to a landing page that corresponds with their rating. Included on that page is a place for them to enter a comment. I’ve gone back and reviewed some of those comments and found some useful trends.

  9. Tim, great post. I think one of the real strengths of a system like this is the fact that you know who clicked what, and therefore you can respond in real time and draw the conclusions that you’ve drawn above. Traditional online surveys make it very hard to work out exactly who is saying what, meaning that you acculmulate more data without actually doing anything about it.

    We (Customer Thermometer) offer a tool that allows anyone to build and send “1-click” style emails similar to the ticket system you have built for yourselves. One of the most consistent pieces of feedback from our customers is that they have actionable data. They can do something immediately with the information coming in because it’s in real time and directly attributable to a specific customer.

    • I agree, the best part of our system, and systems like the one you provide, is the ability to be actionable with a specific customer when it matters.

      I think the overall collection of this data is also important. Knowing the last rating, average rating, number of responses, etc for a specific person when serving them is powerful.

Trackbacks

  1. Why I Treat Customer Service “Experts” With Disdain – Deliver Bliss
  2. Tweets that mention The Truth About Customer Satisfaction | Deliver Bliss - The Business of Customer Experience -- Topsy.com

Leave a Reply