Keep clients coming back · Every business
Measuring Customer Experience Without a Research Team
A small service business can learn most of what a research department would tell it from one short question, a few numbers it already keeps and a monthly hour set aside to read them together.
At a glance
- Customer experience measurement is the habit of asking customers how a visit went, counting what they do afterward and putting the two side by side often enough to notice a change.
- Reaching customers and growing sales was the most common operational challenge in the Federal Reserve's 2024 small business survey, named by 57% of 7,653 employer firms, yet few owners have a steady read on how existing customers feel.
- In a study of customer satisfaction data published in Marketing Science, average satisfaction scores predicted future business performance better than recommendation-based scores such as the Net Promoter Score.
- The practical system is one rating question after each visit, one open comment box and three behavior numbers (who came back, who rebooked, who complained), read together once a month.
- Without it, a business learns about a problem when the appointment book thins out, which is months after customers first felt it.
Part 1 | What you already know is more than you think
Most owners assume that measuring customer experience means hiring a firm, buying survey software and waiting for a report, and so the work never starts. We'd need a research team for that, the thinking goes, and we're a five-chair salon. The assumption puts the hard part in the wrong place, because the data a small business needs is already passing through the front desk every day in two forms: what customers say about a visit and what they do after it.
What customers say arrives as ratings, comments, complaints at checkout and online reviews. What they do shows up in the booking system and the books as the share who rebook before they leave, the share who return within their usual cycle, the average ticket and the customers who quietly stop appearing. A research team would combine those two streams, and an owner can do the same with a spreadsheet and an hour a month, provided the questions stay short and the numbers stay the same from month to month.
The reason to combine them is that each stream has a blind spot. Ratings tell you how a visit felt but not whether the customer will be back, while booking data tells you who left but not why; when a rating drops in the same month that rebooking drops, the comment box usually names the cause.
Part 2 | Choosing the one question to ask
The research on survey scores is more settled than the marketing around them suggests. A study in Marketing Science that tested six satisfaction and loyalty measures against later business results, using American Customer Satisfaction Index data from 1994 to 2000, found that average satisfaction scores had the greatest value in predicting future performance, that the share of customers giving a top score also did well and that recommendation measures, including net promoters, had little or no predictive value. A separate study in the Journal of Marketing, drawing on more than 15,500 interviews about 21 firms in Norway, could not replicate the claim that the Net Promoter Score outperforms other measures in the very industries cited as its best examples.
That evidence does not make the Net Promoter Score useless; it makes it one option among several, and for a small business the right choice depends on what the owner wants to learn. A third measure, customer effort, comes from a study in Harvard Business Review of more than 75,000 customers, which found that 94% of those who reported low effort intended to buy again and 81% of those who struggled to get a problem solved intended to spread negative word of mouth.
| Measure | Question asked | Scale | Best use in a small business |
|---|---|---|---|
| Satisfaction (CSAT) | How satisfied were you with today's visit? | 1 to 5 | Tracking how visits feel month to month |
| Net Promoter Score | How likely are you to recommend us to a friend or colleague? | 0 to 10 | Comparing with other businesses that use the same question |
| Customer effort | How easy was it to book and get what you needed? | 1 to 5 | Finding steps that frustrate customers |
For most owner-run businesses we suggest starting with the satisfaction question, because the research supports it, customers find it plain and it gives a score that moves when something changes. Whatever the choice, the open comment box underneath it matters more than the scale, since a single sentence such as the front desk didn't know my appointment had moved is a fix, while a 4 out of 5 is only a signal.
Part 3 | The monthly scorecard, worked out
The system fits on one page. Send the question by text or email within a day of each visit, keep the comment box optional, then once a month put five numbers in a row: responses, average rating, share of top ratings, share of low ratings and one behavior number such as the rebooking rate. The illustrative month below shows how little arithmetic is involved.
| Visits in the month (Assumption) | 160 |
|---|---|
| Answers received (Assumption) | 48 |
| Response rate (Calculated (48 ÷ 160)) | 30% |
| Ratings of 5 / 4 / 3 / 2 (Assumption) | 34 / 9 / 3 / 2 |
| Average rating (Calculated (219 points ÷ 48)) | 4.56 out of 5 |
| Share rating 4 or 5 (Calculated (43 ÷ 48)) | 89.6% |
| Share rating 3 or lower (Calculated (5 ÷ 48)) | 10.4% |
Illustrative example; replace every assumption with your own numbers.
The five low ratings are where the month's work begins. Read each comment, call or message the customer if they left contact details, and write the cause in a column beside the rating; after three months, the causes that repeat are the ones worth a new step, script or checklist, and the rest are usually one-off bad days.
Two cautions keep the numbers honest. First, the people who answer are not a random sample, so treat a month with fewer than about 30 answers as a hint rather than a trend; second, never tie a discount or reward to a high score, because it turns the measure into a test of how well the front desk asks.
How you compare
- Against national figures: the American Customer Satisfaction Index publishes industry scores built from customer interviews and an econometric model, which differs so much from a one-question survey that the comparison is loose at best.
- Against your own past: the most reliable benchmark is the same question asked the same way last quarter; a drop of a few tenths of a point in the average, sustained for two months, is worth a closer look.
- Against behavior: if ratings are high but rebooking is falling, the survey is probably reaching only your happiest customers, and the gap is the measurement to fix first.
Questions to ask this week
Which single question will we ask after every visit? Pick one and keep the wording fixed so the months can be compared.
Where will the answers land? A shared spreadsheet or the booking system's notes field works, as long as one person owns it.
Which behavior number do we already have? Rebooking rate, return rate within the usual cycle or average ticket are usually in the booking or point-of-sale report.
Who reads the low ratings, and by when? A named person and a two-day window turns a complaint into a recovered customer more often than a monthly review does.
Start by choosing the question and sending it after every visit for one month, then fill in the scorecard above with whatever answers come back.
About this research: This article draws on a Marketing Science study of six satisfaction and loyalty measures using American Customer Satisfaction Index data, a Journal of Marketing replication using the Norwegian Customer Satisfaction Barometer, a Harvard Business Review study of customer effort, Bain & Company's published Net Promoter definition and the Federal Reserve Banks' Small Business Credit Survey. The worked example is illustrative.
Sources
- Morgan, N. A. and Rego, L. L. The Value of Different Customer Satisfaction and Loyalty Metrics in Predicting Business Performance. Marketing Science 25(5), 2006. ACSI data, 1994 to 2000.
- Keiningham, T. L., Cooil, B., Andreassen, T. W. and Aksoy, L. A Longitudinal Examination of Net Promoter and Firm Revenue Growth. Journal of Marketing 71(3), 2007. 21 firms, more than 15,500 interviews.
- Dixon, M., Freeman, K. and Toman, N. Stop Trying to Delight Your Customers. Harvard Business Review, July to August 2010. More than 75,000 customers.
- Bain & Company. Measuring Your Net Promoter Score. Net Promoter System.
- Federal Reserve Banks. 2025 Report on Employer Firms: Findings from the 2024 Small Business Credit Survey. 2025. 7,653 employer firms.
Related reading: How to Ask for Feedback Customers Will Actually Answer · What a Great Customer Experience Is Worth, According to the Research · How Many Unhappy Customers Never Complain, and What That Hides · NPS, CSAT or a Single Question: Which Survey Score to Use · Reading Your Reviews for Patterns: Turning Comments Into Fixes
Work with osNOVA on measuring how customers experience your business
We help owner-run service businesses set up one survey question, a monthly scorecard and the follow-up habits that turn low ratings into fixes, starting from what you want more of, whether that is more regulars, fuller weeks or more first visits that turn into second ones.
Email hello@osnovaops.com about your business
Or leave your email and we will reach out
Tell us what kind of business you run and what you want more of.
Not ready to talk? osNOVA runs free live events for owners.