How Shortcut Metrics Can Pack a Powerful Punch

How Shortcut Metrics Can Pack a Powerful Punch

By Samuel Puchala

Most companies track customer experience (CX) metrics to gain deeper insights into the people who buy their goods and services. But knowing that everyone loves your new product even though they want to strangle your sales staff is only half the battle.

Despite reams of customer data and powerful analytical tools, many of our clients struggle to translate measurement into results. A recent study supports this observation: Although 80% of businesses regularly send out customer surveys, fewer than 10% believe they’re good at using CX to inform key decisions, according to a Temkin Group study.

Companies have plenty of CE metrics to choose from, but finding the right mix that can summarize the performance of your business is no easy task. Many drivers impact the bottom line, so it’s tough to isolate those influences without heavy quantitative research.

Shortcut CX metrics such as Net Promoter Score (NPS), Customer Effort Score (CES), American Customer Satisfaction Index (ACSI), Customer Satisfaction (CSAT), et cetera are popular because they provide a snapshot of performance for a fraction of the work. Reputable studies link these shortcuts to revenue growth even if the science is not perfect.

Yet as broadly applicable and simple to use as they seem, shortcut metrics should be applied with caution. I’ll illustrate best practices using NPS because it has strong quantitative support and is widely adopted. NPS asks one key question: How likely is it that you would recommend a company/product/service to a friend or colleague?

Most of the principles apply to other well-known CX shortcuts. 

Understand what drives NPS

To tie CX to actions, identify what’s driving your metrics and create initiatives accordingly. Articulate and rank the reasons why customers use your product or service. What qualities matter the most to them, and where along the journey do you delight or disappoint?

Answering these questions gives you a reasonable hypothesis when it comes to major drivers. You can use regression analysis to test the effect of such factors on NPS to key in on the most important drivers and discard the red herrings.

Regressing to NPS is much easier than regressing to profit (that’s why it’s a valuable shortcut). During a given period, your financial performance could soar or nosedive for many more reasons than NPS, from wildly successful marketing plans to competitor price promotions.

Identifying all relevant drivers and building a model to validate impact is hard. By contrast, NPS hinges on a smaller and more consistent list of variables. If your NPS results track alongside your bottom-line performance over the long run, managing to this metric will take you in the right direction with relatively little effort.

Tax preparation software specialist Intuit created a predictive model that forecasts NPS based on improvements to its products. The model compelled Intuit to identify NPS drivers, and continually using it forces the California-headquartered company to focus on high-impact changes. Since launch, the model has been accurate within 2 percentage points, and Intuit’s flagship TurboTax product has an industry-leading NPS of 54%.   

Build an ongoing measurement system

Once you’ve established a metric like NPS and developed a detailed understanding of its drivers, institutionalize it using an ongoing measurement system. These systems fall into two buckets: transactional (collecting scores after certain touch points) and periodic (collecting scores through regular random sampling).

Transactional NPS collection provides real-time customer satisfaction data you can use to customize your interactions (e.g. offering discounts to save detractors). You can also assess the incremental NPS impact of individual touch points by comparing your actual post-transaction NPS to predicted NPS. For example, if a customer’s average NPS based on previous touch points was 9 and it fell to 5 after a specific transaction, you can pinpoint the likely culprit.  

Periodic NPS is useful for measuring holistic CX because periodic, transaction-independent measurements will reduce recency bias. It’s also more straightforward. Most businesses already do customer surveys from time to time. NPS measures can piggyback on these studies. There are many powerful tools for automating measurement and analytics. Medallia makes one of the top offerings; we worked with the California-based software developer to design and launch a transactional NPS measurement system for a telecommunications client. The system aggregated data collection and analytics, enabling cross-channel comparison. It also captured data at each customer touch point, delivering transaction-level oversight. The transactional system wasn’t cheap, but it strengthened an important element of our client’s competitive advantage (CX leadership). 

Avoid the pitfalls of NPS

Beyond understanding root causes, there are various technical pitfalls you must avoid when introducing measures like NPS. Here are just a few examples. To put CX metrics into action, you need someone in the organization with the right expertise.

Pitfall #1: Trying to collect NPS from a product or service that no one will recommend  

One of our recent clients has a large and loyal client base. But when we did a preliminary customer survey, we saw bafflingly low NPS. Focus groups revealed that despite high satisfaction levels, customers simply didn’t feel comfortable referring the company’s product, no matter how we asked the question. (It wasn’t Ashley Madison, in case you’re wondering.) For this business, NPS understated true performance and was unsuitable. 

Pitfall #2: Assuming that customer service is the sole driver of NPS  

When searching for NPS drivers, remember that customer service level is one of many candidates. My business partner recently blogged about his disappointing customer experience with Tesla Motors. He’d warn potential buyers about the painful purchase process, but he’s happy with his car and would probably still recommend it. For now, anyway, Tesla has a sufficiently exciting and unique product to overcome some glaring problems.

Pitfall #3: Assuming that NPS always drives bottom-line performance

If you sell a necessary product with few alternatives, you can mistreat your customers and ignore plummeting NPS without taking a financial hit, at least in the short term. Babies “R” Us is a great example. I go there because they’re the only place in town that sells the complicated baby gadgets I need (heavy-duty strollers, titanium cribs). Despite being a high-value customer, I’d sooner recommend discount airbags than this retailer. They’re overpriced and poorly merchandised – and their aisles barely fit the stroller they sold me.

You’ve probably heard Peter Drucker’s maxim “you can’t manage what you can’t measure,” but there’s a big gap between measuring CX and using that data to manage the bottom line. To get results, you need to find the CX metrics that are reliably tied to profit and understand exactly how to go to market in a way that drives them.

Tags: Customer Experience, Measurement, Analytics, Retail