Net Promoter Score...Do customers love you, hate you... or don't they care?

John Camp, Business Applications Consultant - NSC Group

How to reach more of your customers, more often and have a real impact on your bottom line.

I had a discussion with a customer recently about how many interactions they had with their customers. It turned out that they had 3 million interactions per month. From this they surveyed approximately 200 people per month using live agents. This was an attempt to try and understand what their customers thought about the service they offered and obtain a Net Promoter Score. While they were aware it wasn’t nearly enough they weren’t sure what options were available.

So, how do you get good, solid data from your customers with a statistically significant sample size without annoying your customer or breaking the bank?

I have to confess… having been in the contact centre industry (mostly outsourcing) for more than 12 years I really didn’t know a huge amount about Net Promoter Score (or NPS to those in the know) until I moved into my new role last year. Now it’s everywhere, it’s like when you buy a new car thinking hardly anybody owns one and then once you get it out on the road you find that not only does everyone have the same car, they have it in the same colour AND they all know more about it than you do. So I write this blog from a practical perspective based on a recent pilot completed with a large services company in Australia – a steep learning curve indeed and for those who assisted in my education (Sam and Michelle) I thank you!

Firstly, a quick explanation on what NPS is. It is a very simple way of determining whether a customer is a detractor (unhappy with you and scores you 6 or below), an active promoter (tells others about your business in a positive way and scores you 9 or 10) or Neutral (loyal but not promoters and scores you 7 or 8). It does this by asking one simple question:

“on a scale of zero to ten, how likely would you be to recommend [company name] to a
colleague or friend?”

NPS = Promoters - Detractors

Seems pretty straight forward doesn't it? That's one of the great things about NPS, but it doesn't mean it's easy to get it right.

Questions that need to be considered are:

Inbound or outbound surveys
Do we use email, web based, touch tone (DTMF), speech or a combination
What are the advantages and disadvantages of each?
What questions other than NPS are we going to ask and how many?
How do I effectively measure the score against my industry peers

And the most important question to ask yourself:

What is my organisation going to do with the results, how do we increase the number of promoters and decrease the number of detractors?

... and there are a million other questions that need to be answered but you can effectively deploy any measure of NPS.

Given that this is my first blog and short is supposedly sweet I'm not going to try and answer all of these questions but tell you about a pilot that finished in December 2009 for an NSC customer and what they found.

The Pilot

The customer decided on a three month pilot of an Inference natural language speech application. The application was fully hosted, meaning that it could be delivered without the need to deploy any infrastructure in the client environment, this meant rapid deployment was possible.

Inbound surveys were offered at the end of calls by agents in specific teams. The agent determined who was offered a survey although teams were set targets in terms of take up rates - this has significant implications to the survey results.

Something that we believe is a first in the customers industry is that we used a hosted Inference outbound call initiator to do outbound surveys. The application automatically dialed customers that had an interaction with an agent within the last 24 hours and also when the customer had not contacted them within the last 12 months. Those customers were offered the opportunity to take part in the survey or just hang up if they said 'no'.

There were some classic comments made by customers after they had provided an NPS or CSAT score. Customers were invited to provide verbatim answers as to 'Why' they had given the score they had. Positive comments about being treated like a human being when they were experiencing difficulty paying are something that marketing teams love to hear. Negative comments such as having difficulty understanding written communication that was provided to them (after stating that he had a degree in English) again are gold.

Customers were also offered the opportunity to receive a call back from a live agent. This is believed to have stopped frustration leading to a fully blown written complaint and can turn a negative experience into an extremely positive one. You are saying to your customer that you asked, you listed and you acted.

The Statistics

The key to the pilot and to the ability to provide outbound and inbound surveys as part of a single application is volume. It provides the ability to reach every customer, understand how they feel about you and look for themes that you can act upon. The take up rate for outbound surveys reached as high as 8% for those that had recent contact and 5% for those who had not contacted the organisation within the last 12 months. Over a one month period almost four and a half thousand outbound surveys were completed. This means they reached out to approximately 90,000 customers in a one month period. When was the last time your company did this?

From an inbound perspective, over the two month period almost six thousand surveys were completed.

Insights

Some further insights from the pilot were

Customer Satisfaction (CSAT) is largely driven by the service provided by the agent on that particular call - the human factors
It seems that call related issues such as long wait times and IVR issues seem to impact the NPS more than the CSAT
Scores were materially higher for inbound rather than outbound which reflects an agent bias in terms of which callers are offered a survey

More Information

For more information including white papers and ways in which you can effectively deploy NPS please send an email to speechinfo@nsc.net.au