Service Recovery Paradox – Fact or Myth?

This blog is about a well-known business concept called the Service Recovery Paradox (SRP).

What’s SRP? It’s a pretty simple theory. A good recovery following a service failure can turn angry, frustrated customers into loyal ones. Service managers have known about this paradox for years.

The concept was first discussed in a popular Harvard Business Review article as far back as 1990. The authors – Christopher Hart, James Heskett and Earl Sasser – are all extremely well-respected academics and business practitioners.

But are they right? Let’s take a closer look.

The Profitable Art of Service Recovery

Service Recovery at Slack

Why don’t we start with an example.

Ever heard of the American tech company Slack? Several years ago, Slack had a major service failure:

Tuesday, November 23, 2015 started out like any other day at work for James Crenson. Coffee, check email, prep documents for weekly meeting, solicit input from co-workers on slack. Um, Slack?

Outage

At 8:50am, popular workplace messaging service Slack suffered a massive outage, leaving over a million users around the world unable to send or receive messages and files for almost three hours in the middle of the workday. Angry users took to twitter and other social messaging sites to complain about the inconvenience. The situation had the potential to explode, but Slack was ready.

The team messaging tool had a solid plan in place for mass outages. A well-coordinated group effort handled support issues including a comprehensive social media blitz to contain the negative customer experience (CX). Over the few hours that the service was down, the official @SlackHQ account tweeted over 2,300 times with humorous, thoughtful, and most importantly, personalized – responses to customers complaining about the service outage.

Response

Not only did the all-out response wow users, but @SlackHQ gained over 3,300 followers – 7x more than average – on a day that could have gone down as the worst in company history. Slack was able to quickly contain the damage, took complete responsibility, kept its customers well informed and handled a stressful situation with humor and efficiency.

Throughout this process, Slack deepened the trust of existing customers by demonstrating that the company was prepared in times of crisis. Its expert handling of a negative situation enhanced its relationships with existing customers, boosted the brand’s reputation and even served as a springboard for an expanded customer base.

This is an exceptional demonstration of the value of the phenomena known as the “service recovery paradox.”

The graphic below explains the Service Recovery Paradox.

Loyalty generally increases over time when service is delivered consistently well. It falls rapidly if there is a service failure but loyalty increases again when service is restored and the service recovery is handled well. In fact, loyalty becomes greater than if no failure had occurred in the first place. That’s the paradox.
Service Recovery Paradox

Fact or Myth?

The Service Recovery Paradox is generally accepted as fact. But where is the evidence for it?

Some years ago, Celso Matos decided to find out. He and a couple of colleagues from the Federal University of Rio Grande do Sul in Brazil conducted a ‘meta-analysis’ of all academic articles that discussed SRP. In total, they found 24 documented examples of recoveries following service failures. 19 of the 24 examined the impact of service recovery on satisfaction; 12 examined the impact on repurchase intentions; six looked at word of mouth (advocacy).

Their results were very interesting and not encouraging for companies with a poor service ethos.

Matos concluded that “satisfaction increases after a high service-recovery effort” but that “repurchase intentions are not increased by a high service-recovery performance.”

Service Recovery does NOT lead to greater loyalty

Matos and his colleagues explain that “customers are willing to make a positive evaluation of a firm providing a high recovery effort, but they are not likely to repatronize this firm”. They go on try to explain why this might be the case.

One explanation is that satisfied customers are not necessarily loyal. We know this from our own analysis at Deep-Insight. Another explanation is that people will give you a lot of credit for pulling out all the stops to recover a bad situation after a bad service failure. However, they will still doubt your ability to ensure no similar service failures occurs again. That’s pretty important when your 12-month or 10-year contract is coming up for renewal. They might give you good customer satisfaction (CSat) scores, but will they sign up for another contract?

In general, satisfaction levels do recover if a Service Improvement Plan (SIP) is put in place but loyalty does not. This does not mean the Slack example discussed above is not true. It just means that you need to go to extraordinary lengths to win back the trust and loyalty of clients when they experience a service failure.

The Relationship between Service and Trust

Some years ago I wrote a blog called Trusted Relationships = Consistently Good Service. It was based on our analysis of tens of thousands of customer feedback results from Deep-Insight’s database gathered over a period of 15 years. The analysis showed that of all the elements in our Customer Relationship Quality (CRQ™) methodology, Service Performance was most strongly correlated with Trust.

It stands to reason, if you think about it. What happens when you deliver a consistently good service to a client? Their levels of trust in you and commitment to your company increase. Fail to deliver the service consistently, and their trust erodes quickly. Consistently fail to deliver, and both trust and commitment levels can disappear completely. Trust and commitment also take a very long time to rebuild.

‘Quality is Free’ so make sure to get it ‘Right First Time’

There’s a really important lesson here. Service failures are the sworn enemies of long-term profitable relationships. That’s why it’s worth investing time and effort to minimise the chances of a service failure ever happening in the first place. You may never succeed completely but it will be worth the investment. In fact, it won’t cost you anything.

This is where the old Total Quality Management (TQM) principles come into play. In 1979, Philip Crosby wrote a seminar book called Quality Is Free. His basis message was that it costs absolutely nothing to build quality into products and services. If anything, you’ll make money by reducing the cost of re-work and failed business relationships.

Quality really is free, so if you’re in the service business, make sure to get it right first time.

If you operate in the B2B world, ask yourself the question: “Is CSat the right thing to be measuring in the first place?” Maybe you should be measuring something different.

Are you going to NPS me? Yes, I am!

This is the topic of a talk I’m giving this week at a customer loyalty conference in Melbourne. It is in response to another talk entitled “Are you going to NPS me? No I’m not” in which Dr Dave Stewart of Marketing Decision Analysis will be presenting the case that Net Promoter is a deeply flawed concept. Dave will say that NPS should be discarded by organisations that espouse customer advocacy. To be honest, Dave’s position is probably close to what I thought of the Net Promoter Score concept when it was first introduced by a pretty smart academic and business consultant called Fred Reichheld back in 2003.

Tom Fishburn

Net Promoter Score

Reichheld’s basic premise was that you only need to ask one question in order to understand if a customer is going to stay loyal to you or not. The question is: “How likely are you to recommend us to a friend or colleague?” Fred, being the excellent marketeer that he is, proclaimed the benefits of this Net Promoter Score (NPS) concept in respected publications like the Harvard Business Review. He then promoted it in his own book The Ultimate Question which came out in 2006, shortly after I took on the CEO role here at Deep-Insight. Since then, NPS has became very popular as a customer loyalty metric.

However, NPS has also attracted some heavy criticism. Tim Keiningham gave NPS a particularly scathing review saying that he and his research team could find no evidence for the claims made by Reichheld. (It should be said that Keiningham worked for the market research company Ipsos so his views may not be completely unbiased.)

At that time, my own view was that NPS was probably too simplistic a metric for business-to-business (B2B) companies. I also felt that Deep-Insight’s own customer methodology – which also included a ‘would you recommend’ question – was a much better fit for complex business relationships. And if I’m honest, there was an element of ‘Not Invented Here’ going on in our own organisation as well.

So we decided to ignore NPS.

The Rise of NPS

But here’s the thing: our customers didn’t ignore it. When we ran customer feedback programmes for customers like Reed Elsevier and Atos in the UK, ABN AMRO in the Netherlands, Santander in Poland, and the Toll Group in Australia, they would all ask: “Can you add in the NPS question for us – we have to report the numbers back to headquarters?” Of course, being the good marketeers that we were, we duly obliged. However, we always gave the results back in a separate spreadsheet, so that it wouldn’t contaminate our own reports and our own wonderful methodology!

Roll the clock forward to 2013. NPS still hadn’t gone away. In fact it had become even more popular, particularly with large international companies where a simple understandable metric was needed to compare results across different divisions and geographical areas. And when I finally looked into it, I discovered that Deep-Insight had actually been gathering NPS data from customers across 86 different countries since 2006.

Is NPS a good predictor of loyalty?

Around the same time we also did some research into our own database to find out what really drove loyalty and profitability in our clients. Now this is not an easy thing to do, as many of you who have tried will know. But where we had several years of customer feedback data, it was relatively straightforward to analyse how many of our clients’ B2B customers were still with them. If they have deliberately defected, we investigated if that defection could have been predicted by a poor Net Promoter Score, or by any of the metrics in our own CRQ methodology.

I have to say that the results were quite interesting. A low ‘Likelihood To Recommend’ was not the BEST predictor of customer defection. However it was actually a pretty good predictor. Deep-Insight’s overall Customer Relationship Quality (CRQ) metric was a slightly better predictor. A poor Commitment score – one of the key components of CRQ – was the best predictor of whether a B2B client was going to defect to the competition or not.

So there we had it: NPS did actually work.

It worked not because it’s the BEST predictor of whether a client was going to defect, but because it’s a GOOD predictor, coupled with the fact that NPS has been embraced by some of the world’s leading organisations as an easy-to-use and internationally-accepted customer benchmark. At Deep-Insight, we came a little late to the party. We only incorporated Net Promoter Score into our customer methodology in early-2014. Today we find that the combination of NPS and our own CRQ metrics works really well for our clients.

The future for NPS

Now let’s go back to the cartoon at the top of the blog (and thank you to the wonderful Tom Fishburne for allowing us to use it). If there is a statistically purer methodology than NPS, why not use that instead?

The answer is simple: most senior executives aren’t interested in re-inventing the wheel. They are much more interested in taking the feedback from their clients and acting on it, so that they can protect and enhance the revenues they get from those clients.

So for those B2B executives who are wondering if NPS is the right customer metric for them or not, I would suggest that you’re asking the wrong question. What good CEOs and Sales Directors are asking these days is:

“If my Net Promoter Score is low or if I have a lot of Opponents and Stalkers as clients, what do I do?”

In fact, the really successful CEOs and Sales Directors are spending the time thinking about the challenges of putting a really effective customer experience (CX) programme in place, rather than worrying about the purity of the metrics. That’s what you should be doing too.