Is NPS the right measurement to understand your business performance?

Let’s talk about Net Promoter Score for a moment, shall we? First, let’s just make sure we’re all on the same page – a definition:

NPS, at its core is a calculation of customer responses to a question, typically something like “How likely are you to recommend <thing> to a friend or colleague?” with answers being on a 1-10 scale.

To get your NPS score, you take the number of “promoters” – those rating either a 9 or 10 – and subtract from it the number of “detractors” – those rating at 6 or below. You then end up with a score somewhere between -100 (all detractors) and +100 (all promoters).

So, there we go. Simple. Contact your customers, ask them how likely they are to recommend you, do the maths and wear your NPS score like a birthday badge if it’s good, or bury the results and make lots of announcements about how you are now “seeking to be a customer-focused organisation” if it’s bad.

Good work everyone, double espressos all around, back to the day job.

Or not. Hopefully. To really understand the experience you’re offering to your customers, is NPS enough? Can it be more than a headline number? And, if not, what are the metrics that really matter when trying to understand your customers’ experience of you.

For starters, it’s worth acknowledging that NPS can be measured for different things within a company. The Brand level is where many organisations focus and where many industry reviews will begin and end. At this level, NPS probably is little more than a birthday badge – who has the biggest number; whose is the most popular brand.

Transactional NPS works at the level of an individual interaction. As a customer you’ll have experienced this after an online chat request, when your support agent asks you to complete a survey asking whether they resolved your issue. At this level, you can start to build an impression of which interactions and capabilities work well in your organisation.

But At any level, NPS is not without problems. A few observations:

  • Are all 8/10s equal? For me, 8/10 may be a really good company – one that I’m very happy with. After all, 80% is an A-grade in most examinations. Everyone has room for improvement somewhere, but you’re good enough. Yet, according to the method, this only makes me a “passive”; I don’t contribute to your NPS score. Catch me on a bad day (one which may not even be your brand’s fault) and I might be inclined to only rate you 6/10, even for the best outcome. That makes me a detractor. So, my view of why I have given you the score I did will be personal to me and influenced by my context and circumstance
  • What you mean to me may also influence how I response to the NPS question. Honestly, if I find myself in a position where I’m talking about mobile phone packages, or haemorrhoid cream or accounting software, let alone having to make a recommendation to a friend or colleague, I’m going to want to re-examine my choices in life. So the fact I’m scoring you 5/10 doesn’t mean I don’t like your product or service, it just means I don’t want to be the sort of person who discusses the relative merits of Wifi extenders.
  • Why do I like/dislike you? And here we start to reach the core of the problem with NPS: What is my reason for giving you the score I did? Maybe your product is OK, but your customer service is bad. Or you’re a good company but your products are just too expensive (or a lousy company with cheap products).
  • Context is king. If I’m buying a product, especially one that I’m excited about like the brand new smartphone, I’m likely to be really happy with any sales process that lets me buy it, unless you really screw it up. If I’m querying an error on my 5 page phone bill, or trying to find out why my “upto 40Mb” broadband line is only getting 17Mb, I’m unlikely to be delighted with the outcome unless you can work some magic.

You could argue that these factors balance out: the people having a bad day and marking low are cancelled out by those having a good day and marking high. Or you could say that these factors will apply equally to every company, so you still end up with a valid way of comparing your company’s performance to your competitors’.

However, what you do not get out of an NPS score is a good vision of the business functions in your company that need to change and how they need to change to make you an industry leader, or even a better company.

Coming up in part 2… If not NPS, then what should we be measuring?

Setting the Right Expectations – a CX Cornerstone

The route to the top of Customer Experience in the Broadband market, especially in the UK, has been fought, as much as anything, over headline speeds for downloads as any other feature.

BT offer an 80Mb speed, Virgin trump it with 100Mb. BT counter with 150Mb, Virgin respond with 200MB and so on, seemingly ad nauseum.

In some UK telcos, complaints or queries over broadband speed can comprise upwards of 40% of the inbound calls and queries from customers. Only billing rivals speed as a source of customer dissatisfaction; all other issues are some distance behind.

And, to be honest, it’s not really surprising. With most other goods and services, you expect a correlation between what you’ve been sold and what you get. Buying a backpack or suitcase advertised as “holding upto 80 litres” only to find it would struggle to hold half that volume would lead to complaints, so why not a broadband service that claims to offer “upto 80Mb” but in practise only gives 40Mb?

From a technical, network perspective there are dozens of reasons why broadband speeds vary from house to house, let alone country to country and the purpose of this blog is not to explore them in depth. Suffice to say that – for the most part – in countries such as the UK which rely predominantly on copper connections, the longer the wire – the slower the speed. And the more people using the network the slower any one person’s experience of it will be.

New rules are coming in which will seek to address this head on but, as with many other technology-related laws it remains to be seen whether the intention matches the outcome. I’m sure that nobody involved in the legislation regarding cookies used on websites intended to leave every internet user plagued with notifications taking up the screen every time they visit a website.

There are two key lessons for fixed line telcos to learn here. One is, perhaps, more obvious than the other. The more straightforward-sounding lesson is to be more honest about what the customer is being sold.

The new legislation will help towards this. It is obliging CPs to move away from selling broadband based on a headline “upto” speed and towards an average speed. There will be a formula that CPs have to use to determine what figure they can use for this average speed – based on a distribution curve of what customers currently taking the product receive today; speeds achieved during quiet periods ( weekday mornings) and those achieved during the busy hour (typically early Friday evenings) for customers of the same type as them.

It should go without saying that a customer sold a broadband package on the basis of “average 40Mb download” is likely to be happier with the purchase if he’s getting 40Mb, than if he bought the same thing when sold as “upto 80Mb download” even if everything apart from the name is identical between the two. Set an expectation with the customer that you can, and will, meet is hardly rocket science in the world of CX.

Dig a little further beneath this, though, and things will get a little more complex. Especially in a market like the UK which has such a mix of technologies delivering broadband; where some customers are still stuck on technologies that cannot deliver more than 8Mb, others 16Mb, still others on 24Mb, right the way through to the annointed few with full fibre (FTTH) connections who can get hundreds of MB.

Even worse, millions of premises will be served by more than one of those technologies, meaning someone could be paying £40+ per month for an “upto 24Mb” service when a cheaper, faster service is already available to them. And, to complicate matters further, for some people their speed on the faster service may actually be lower than their actual speed on the slower, older technology.

Still with me? Phew!

So, confidence that this new legislation will guarantee a country full of happy customers, who are getting exactly the broadband service they want and need is, understandably, not quite at 100%.

Maybe it should go a little further? Gas and electricity suppliers have recently been obliged to start telling existing customers whether they are on the cheapest tariff. Maybe telcos should be obliged to provide clear options to customers about the services available, a realistic view of the performance they are likely to recieve, and their cost? For example

  • You are currently paying £40/month for our 24Mb service and you are getting an average of 21Mb
  • You could switch to our £25/month service where we predict you will get an average of 38Mb
  • Or our £30/month service where we predict you will get an average of 72Mb

Certainly, most telcos will have these facts to hand, based on their own knowledge of their network and its performance.

However, this leads to my second suggestion on how to improve this experience. What do these numbers mean, to ME as a customer?

Based on the example, above, I can see that it’s silly for me to be paying more for a service that gives me less. But what does 38Mb actually mean to me? Do I need that extra 34Mb, or will I just be spending £5 a month I don’t need for a 72Mb product?

If you are a quadplay telco, your range of products and options for customers to choose will already be hellishly complicated – different TV packages, sport or no sport, HD, SD, Netflix, on demand, IP calls, PSTN calls etc etc. Asking customers to also choose whether they want 38 or 72 (or 150 or 300 or whatever the future offers) is an unnecessary complication.

Bandwidth does not need to be a primary differentiator. In itself it does not do anything for your customer, apart from a tiny bragging right to their mates, for some people. For the rest of us, bandwidth is a means to an end.

Living on your own, or not interested in streaming ultra-high definition content? You’re unlikely to need more than 16Mb. Ever.

A switched-on family with kids at home? Likely to need to support multiple HDTV streams and HD sport and gaming (and uploading a Twitch stream) and phonecalls? Much below about 60Mb is going to mean someone in the household will miss out.

So maybe the smart way to resolve the customer experience around broadband is to stop making the broadband customer experience about speed and move it to services. In the immediate term, this would remove issues around customers who are paying for an “upto 80Mb” service and only receiving what they’d get from a cheaper “upto 40Mb” service – and eliminate potential cases of mis-selling.

In the longer term, this gives customers a simpler buying experience and, in all likelihood, a simpler in-life service experience. Ignore whether I’m getting 40, 80, 150Mb and focus on whether I’m consistently getting the services I’m paying for at the quality I expect:

  • Do all my Netflix films stream in HD?
  • Does my football match, or grand prix buffer?
  • Does my audio download jitter?
  • Are my phonecalls clear?
  • Do my online games lag?

If the answer is always yes, your customers won’t care what their download speed is and customer satisfaction should improve. Sustainably.

New Services, new customers – a new, old challenge for the telco industry

The comms industry faces a variety of challenges to remain profitable – against a background of falling mobile handset sales and the move away from traditional calls revenue towards IP voice services. The key defence against this is by growing new revenue streams through enablement of new services to customers. And the common feature that these services need is higher bandwidth links to the customer.

Providing more bandwidth will invariably require some uplift in physical comms infrastructure, whether in the exchange, mobile mast deployment, roadside cabinets, the cables between them or the cables out to the end customer’s premises. Delivering this infrastructure is an expensive, complex process typically requiring (in the UK at least) coordination with Government agencies and numerous third parties. While the model may differ slightly between countries with an established fixed-line network and those which are more mobile based, similar challenges can be found.

As well as being the means to increased revenue, infrastructure build is a very expensive activity and, when managed poorly, one which can lead to a high degree of wasted investment. There are two headline metrics for this:

  1. The cost per end customer to deliver new infrastructure. This is typically around £400-500 per customer, but can be as high as £700-800
  2. The %age of spend which ends up in the ground, delivering services, as opposed to being wasted on management, rework or other cost of failure. Again, while the average is around 40-60% of investment actually being spent on infrastructure, some companies struggle to reach half this figure.

The efficient delivery of infrastructure depends on a well managed lifecycle, to:

  • identify the demand within a particular geography (a street, housing estate, exchange area)
  • prioritise the areas of high demand – those which will generate the greatest return on infrastructure investment
  • understand the “bill of materials” needed to deliver working infrastructure end-to-end
  • manage the build of this infrastructure to an efficient and timely completion
  • bring the new infrastructure into active use and be able to sell products and services to end customers, through it
  • maintaining an ongoing view of existing uptake and future demand to feed into the build plan

While it may be simple to think the customer’s concerns focus on straightforward purchasing a service once the infrastructure to support it is available, brand reputation and customer satisfaction can be impacted much earlier than this. Chief amongst these Moments That Matter to the customer will be a lack of communication during the build phase, especially over any delays that might arise which will impact the date when the promised service will be available.

At this stage, the customer’s expectations will be high. They will have been contacted by the telco to identify their interest in new services. They may have been offered promotional rates, as an inducement to sign up early to the new services. They will believe that any dates communicated to them are committed. As a bare minimum, it will be necessary to provide regular updates to the customer, reassuring them that work is progressing according to plan. In situations where delays arise, customers will value honesty and transparency over this with early visibility of issues and changes to plans.

As more companies move into infrastructure build, within the UK at least, customer experience will be impacted by a wider number of moving parts. Some telcos may take direct ownership of all aspects of the plan and build; some telcos will outsource some, or all of their plan and build activities to one, or more, third parties; while others will remain in the retail-only space, reliant on infrastructure built by other organisations.

Those companies managing, or controlling their own infrastructure build will want to ensure as much of their investment goes into service-delivering infrastructure to maximise return on investment and reach the greatest possible footprint.

In all of these scenarios, it will be vital to provide customer experience differentiation, through a tightly integrated stack which allows marketing data to drive planning decisions; planning to be efficiently allocated to labour teams; the progress of those labour teams to be closely monitored, optimised and reported on, both internally and to customers; and for revenue income to be maximised by efficient sales enablement.