Short answer
It is easy to measure revenue and much harder to measure trust, but trust is usually the more important number in a relationship-driven business. Revenue, transaction count, and website traffic are simple to track and naturally dominate attention, precisely because they are easy to put on a dashboard.
Easy-to-measure numbers can crowd out important ones
Revenue, transaction count, and website traffic are simple to track and naturally dominate attention, precisely because they are easy to put on a dashboard. Harder-to-measure things — customer trust, referral quality, long-term relationship health — get less attention by default, not because they matter less, but because they resist easy quantification.
We try to deliberately correct for this by building rougher, imperfect proxies for the harder-to-measure things rather than ignoring them simply because a clean number does not exist.
Referral rate is a better trust signal than satisfaction scores
A satisfaction survey captures how someone felt about a specific transaction. Whether they refer a friend or family member afterward is a stronger, more costly signal, because a referral puts the referrer's own credibility on the line. We watch referral patterns closely, more closely than we watch survey scores, as a more honest read on whether we actually delivered.
Referral rates are also harder to inflate through small transactional gestures, which makes them a more durable metric than most satisfaction measures over time.
Repeat relationships reveal more than first transactions
A first purchase or first matchmaking engagement tells you whether marketing and first impressions worked. A second one, freely chosen without any special incentive to return, tells you whether the actual experience delivered on what was promised. We weight repeat engagement heavily in evaluating whether a part of the business is genuinely working.
This is a slower signal to gather than first-transaction data, but it is far more reliable, and it has occasionally told us that an area of the business needed attention well before more immediate metrics would have flagged it.
Track the cost of quality failures, not just their frequency
Counting how often something goes wrong is useful, but understanding the actual cost of each failure — in time, money, and relationship damage — gives a clearer sense of where to invest in prevention. A rare but severe failure can matter more than a frequent minor one, and treating them as equivalent misallocates attention.
We try to weigh issues by real impact rather than simply by frequency when deciding where to invest additional quality control effort.
Employee and partner sentiment deserves regular, honest measurement
It is easy to assume things are fine internally simply because nobody has complained loudly. We try to check in more actively — direct conversations, not just an annual survey — with staff and close partners, because quiet dissatisfaction tends to surface as attrition or reduced effort well before it surfaces as an explicit complaint.
This kind of measurement requires genuinely wanting to hear uncomfortable answers, which is harder than it sounds when things otherwise appear to be going well.
Long time horizons change what counts as a good number
A metric that looks good over one quarter can look very different over five years, particularly anything related to trust or relationship health. We try to evaluate key metrics over multi-year windows where possible, not just the most recent reporting period, because the most important effects in this kind of business play out slowly.
This longer view has occasionally led us to stay the course through a metric that looked mediocre in the short term but was consistent with a longer-term trend we had more confidence in.
What to remember
- Build proxies for hard-to-measure things like trust rather than ignoring them.
- Weight referral and repeat-engagement rates heavily; they are harder to fake than surveys.
- Evaluate quality failures by real impact, not just frequency.
- Check in on staff and partner sentiment actively rather than waiting for complaints.
Frequently asked questions
- What should you know about easy-to-measure numbers can crowd out important ones?
- Revenue, transaction count, and website traffic are simple to track and naturally dominate attention, precisely because they are easy to put on a dashboard. Harder-to-measure things — customer trust, referral quality, long-term relationship health — get less attention by default, not because they matter less, but because they resist easy…
- What should you know about referral rate is a better trust signal than satisfaction scores?
- A satisfaction survey captures how someone felt about a specific transaction. Whether they refer a friend or family member afterward is a stronger, more costly signal, because a referral puts the referrer's own credibility on the line.
- What should you know about repeat relationships reveal more than first transactions?
- A first purchase or first matchmaking engagement tells you whether marketing and first impressions worked. A second one, freely chosen without any special incentive to return, tells you whether the actual experience delivered on what was promised.
- What should you know about track the cost of quality failures, not just their frequency?
- Counting how often something goes wrong is useful, but understanding the actual cost of each failure — in time, money, and relationship damage — gives a clearer sense of where to invest in prevention. A rare but severe failure can matter more than a frequent minor one, and treating them as equivalent misallocates attention.
- What should you know about employee and partner sentiment deserves regular, honest measurement?
- It is easy to assume things are fine internally simply because nobody has complained loudly. We try to check in more actively — direct conversations, not just an annual survey — with staff and close partners, because quiet dissatisfaction tends to surface as attrition or reduced effort well before it surfaces as an explicit complaint.