Having seen a number of articles/heard a resurgence of opinions recently arguing for/against different CX metrics yet again, it’s time to clear my throat and assertively scream once more.

The best CX metric is that which best explains the customer behaviors you are trying to motivate and the business outcomes you want to drive.

This isn’t mere metric agnosticism. I’m not shrugging my shoulders and saying it doesn’t matter or I don’t care what metric is used. It’s empiricism. The best metric can only be determined by testing what explains or predicts the outcomes that create value for your firm.

Everyone wants measurements that matter – there isn’t much point in having metrics that fail to or inaccurately measure what you are trying to accomplish. In fact, measuring the wrong things is both a waste of resources and can prompt misguided decisions that undermine corporate objectives.

The strategy behind boosting CX performance is that it creates value for the firm. The protestations of the apostles of NPS, CES, OSat or what-ever-CX-metric notwithstanding, the best CX metric depends on how your company creates and measures value, which will depend on your company’s business model, customers, marketplace, distribution channels and competitors. Which metric does this best is something that can (and should) be readily tested.

Read More: The Future of Customer Engagement

CX Objectives

The core value proposition for investing in CX is that it helps companies sell more stuff. While not mutually exclusive, this usually means focusing on

• Customer retention (to continue to sell stuff to existing customers)
• Cross-selling and upselling (to sell more/other/more expensive stuff to existing customers)
• Getting customer referrals and recommendations (to have current customers help motivate others to buy your stuff)
• Capturing a larger share of customer spend
• Or whatever else creates value for your company.

The steps for determining the optimum metric are straight-forward

1. Identify the outcome(s) that are the most important to your business,
2. Link the outcomes to customer survey and behavioral data and
3. Test which metric does the best job of explaining those outcomes.

That is your North Star metric for driving the business forward, it’s connecting the experiential data with the Outcomes (or Operational) data to quantitatively determine the best CX metric for your company.

Many companies, however, skip this step and blindly follow a CX metric that they inherited, assumed was “right” or was imposed on them by someone more senior. Then they wonder why the metric isn’t explaining any meaningful business outcomes or, worse yet, the metric moves in one direction, the business in the other. They then often assail the measurement process: the numbers are wrong! In all likelihood, however, it’s not that the measurement is wrong, but that they have been measuring the wrong thing.

There Always Are Caveats

You will need to prioritize the outcomes you want to drive, as you statistically cannot simultaneously maximize every objective. Pick the ones that are most important for your business. Some outcomes, moreover, are more difficult to quantify than others, so you may need to be creative or rely on indirect measures.

Most firms tend to focus on retention – and for good reason. For the large majority of companies, 90+% of their revenues next year will come from this year’s customers. This places a huge premium on retention. For companies that focus on customer lifetime value (CLTV), moreover, what better way is there to boost lifetime value than to extend the lifetime of the customer? If your business model is based on infrequent large ticket sales – say selling major appliances or real estate – retention is less important, as you might starve before a customer comes back to buy another fridge or house. In the former case, cross-selling other products may make more sense as a metric, while the real estate model is more likely driven by referrals.

The time lag between stimulus and response also is a factor. That is, how long after a great/lousy experience does the customer respond? There is no simple answer here, but be aware that the experience/reaction dynamic usually isn’t instant and there might be a lag of three or six months or a year or even longer, especially if there are contracts in place and in B2B settings.

Read More: How Direct-To-Consumer Brands Can Build a Superior Customer Experience

The Political Wild Card

Proof, of course, doesn’t always carry the day. While working on the client side I brought my data to the CEO showing that a certain set of variables explained the allocation of market share in our duopolistic B2B business, the key metric that drove our profits. The CEO (a Harvard-trained Ph.D. Economist) dismissed the data, saying that it simply indicated that our customers weren’t smart enough to know what should drive their decisions… and we continued to measure what the CEO said should be our customer’s decision criteria.

Politics are a reality. To this day, I continue to work with clients who are obligated to report on a corporate metric that has been postulated even when other metrics have been proven to be important to the business. They then simultaneously run the numbers behind the scenes on the metrics that really drive the business forward and quietly work on the metrics that matter.

Not ideal, but practical.

Measure Twice, Cut Once

I’m not much of a handyman, but there isn’t much point in aiming for anything less than accurate measurement…otherwise, why measure at all? The metric, keep in mind, is not the objective, but an indicator of the objective.

None of the popular CX measurements are inherently bad: they all are intercorrelated and will point in the same general direction. None of them will lead you to shoot in a direction that misses the barn, but some will be closer to the bullseye than others.

We know why we measure: so we can manage and make better-informed decisions. If you accept the premise that customer experiences matter and influence their subsequent buying behavior and that customer behaviors can either create or destroy value for your firm, then it’s clear that you should determine which CX metric best explains the behaviors that create value for your company. That, after all, is the best CX metric for you.

Read More: Real-Time Customer Experience Is Happening, with or Without You