31 Aug Matchmaker Success Rates: How to Read the Claims
Somewhere in your research you have run into a figure. A firm reports that some large share of its clients found what they were looking for. The number is round, it carries no footnote, and it is built to end a conversation rather than start one.
A percentage is not a fact. It is the output of three separate decisions: what counts in the numerator, who counts in the denominator, and when the count is taken. In this category all three are made by the company selling the service, and no outside body checks any of them.
That is not an accusation that every firm publishing a number is lying. It is a warning that you cannot tell which, since two firms reporting identical figures may be measuring completely different things and neither is obliged to say which. What follows takes the three decisions in turn, then offers better questions than the one you came here to ask.
A percentage is three decisions, not one fact
Any rate is a fraction with a clock attached. Change any one of the three decisions and the same underlying reality yields a modest figure or an impressive one, without anybody writing a false sentence.
Where measurement is taken seriously, those decisions are constrained by somebody other than the party being measured. Here they are not. California regulates this business, but it regulates the paperwork. Under California Civil Code section 1694.2, a dating service contract must be in writing, the buyer must get a copy at signing, and the cancellation notice must appear in at least 10-point boldface type near the signature line. Real protections, and entirely silent on how a firm may define, calculate, or advertise an outcome.
Advertising law reaches further. Under the Federal Trade Commission’s endorsement rules, an ad relating consumer experience on a key attribute of a service will likely be read as representing what customers generally achieve, and the advertiser must possess adequate substantiation for that representation. The rule also notes that disclaimers along the lines of “results not typical” have been tested and do not cure the impression. That standard is enforced after the fact, by complaint. Nothing in it puts a figure in front of an auditor before it goes on a website.
The numerator: what counts as a success
Start at the top of the fraction, where the widest gap sits. A firm could count a second date. Or a third. Or a relationship that reached exclusivity, or one that lasted a year, or a marriage. Or a client who reported being satisfied with the introductions whether or not any of them lasted. Or a client who renewed, or referred a friend. Each is a defensible reading of the word success, and each produces a wildly different number from the same set of clients. Then there are the awkward cases, like the couple who dated eight months and then ended it.
When one organization both defines the outcome and reports on it, the definition drifts toward whatever is easy to reach. The clearest documented example comes from another industry. In 2014 the Consumer Financial Protection Bureau sued a for-profit college chain, alleging that the schools counted a job lasting one day, with the promise of a second day, as a career, and further inflated advertised job placement rates by paying employers to hire graduates temporarily. Those were allegations, not proven findings, and the point is not that matchmakers behave this way. It is structural. Placement rate sounded objective until somebody read the definition underneath it.
So the question is not what a firm’s rate is. It is what the smallest event was that counted toward it.
The denominator: who gets counted
The bottom of the fraction is quieter and does more work.
Everyone who signed? Everyone who completed the full term? Everyone who accepted a minimum number of introductions? Everyone who, in the firm’s judgment, followed the process? Each population is smaller and more flattering than the last, and the client who left in month two, exactly the person whose experience you want, disappears along the way.
Medicine handled this by naming it. In clinical trials, intention-to-treat analysis includes every subject who was randomized, regardless of adherence, protocol deviation or withdrawal, precisely because removing the people who did not comply produces overoptimistic estimates of how well a treatment works. Per-protocol analysis counts only those who finished. Both are legitimate, because trials state which population they analyzed and reviewers check. A matchmaking figure states nothing, and nobody occupies the reviewer’s chair.
There is also the screening effect. A firm that declines most applicants will report better outcomes than one taking nearly everybody. Part of that is real skill in judging readiness. A good share is arithmetic. A rate produced by heavy selection at the door describes who was turned away as much as the work that followed.
The window: when the count is taken
The third decision is when to stop the clock, and almost nobody discloses it.
Measured at the end of a contract term, a figure captures early enthusiasm. Measured a year later it captures something more durable and gets far harder to collect, since it means staying in touch with people who have no reason to answer. Measured five years out it would be genuinely informative, and essentially nobody does that.
Compare how careful measurement handles the same problem. The National Center for Health Statistics report on first marriages states plainly that its estimates describe the probability that a first marriage will last up to 20 years rather than for a lifetime, and warns readers that its longer-duration estimates are biased toward people who married at younger ages because of how the sample was drawn. That is a declared window, with an admission of the bias the authors could not remove. Set it beside a round number with no date attached.
Survivorship runs underneath all of it. Outcome data here comes from clients still reachable and still willing to reply. People whose search went badly stop answering. They are the missing part of every denominator, and no firm reporting on itself recovers them.
What happens once a number becomes the goal
There is a further problem, and it is not about honesty. It is about what publishing a figure does to the people publishing it.
When New York and Pennsylvania began releasing cardiac surgery report cards, economists studied what changed. Using national data on Medicare patients at risk for cardiac surgery, they found that the report cards led to selection behavior by providers and, on net, to higher resource use and worse health outcomes, particularly for sicker patients. The surgeons were not falsifying anything. They were responding rationally to being scored, and the rational response was to take on fewer difficult cases.
Apply that here. A firm marketing on an outcome rate has a standing incentive to sign the searches most likely to go well and to be discouraging about the rest. The people most easily steered away are those whose requirements are narrow or who are further from ready. If your search is complicated, a published rate is not neutral information about a firm. It describes an incentive that may work against you.
The same dynamic shows up in reviews. The endorsement rules cited above treat it as potentially deceptive to solicit public reviews only from customers who already gave positive private feedback. Curating a review page and curating a success rate are the same act with different tools.
Better questions than “what is your success rate”
Ask for a rate and you get a number and are no wiser. Ask these instead, and listen to how readily they are answered.
How do you define a successful outcome, and what is the smallest result that counts? Anyone who has thought about it has an answer ready.
Who is included when you describe your results? Everyone who signed, or only those who finished the term? Ask what happens, in the arithmetic, to a client who leaves early.
When do you stop counting, and how do you stay in touch that long? The second half separates a real practice from a marketing line.
What do you track internally that you do not publish? Renewals, referrals, how often an introduction leads to a second meeting. A firm keeping honest records describes them without much prompting.
What kinds of searches do you decline, and why? The answer shows the firm’s judgment and the selection running underneath anything it reports.
Fee structure is a separate conversation, covered in what drives the cost of a matchmaker. If you are still working out what the service involves week to week, start with what a matchmaker actually does.
A note on how we work
Los Angeles Singles has been introducing relationship-minded people in this city for more than 25 years, from offices in West Los Angeles and Encino.
We do not publish a success rate. Not a flattering one, not a cautious one, not one with an asterisk. We have no figure verified by anyone outside this company, and we will not ask you to weigh a number we defined, calculated and audited ourselves. It would also be incoherent to explain at length why these figures are not comparable and then hand you ours.
What we point to instead is imperfect, and we would rather name the imperfection than dress it up. Our member feedback page carries first-party reviews, many naming the matchmaker the client worked with, and our success stories describe particular introductions that led somewhere. Those are accounts, not measurements. They are self-selected, we host them, and people whose search disappointed them rarely sit down to write one. Read them as evidence of what the good version looks like, not of how often it happens.
We are one credible option among several in this city. Every question above works on us, and we would rather you ask it here than skip it somewhere else.
Frequently asked questions
What is the success rate of matchmakers?
There is no comparable figure. Published rates are defined and calculated by the firms that publish them, and are not independently audited. One firm may count a second date, another an exclusive relationship, another a marriage, another a client who said she was satisfied. Each also picks which clients to include and when to measure. Ask instead how a firm defines success, and who is counted.
Why would a matchmaker refuse to give me a number?
Two very different reasons look alike from outside. The respectable one is that the firm has no substantiated figure and declines to invent one, which is our own position. The unrespectable one is that its outcomes are poor. Tell them apart by asking what the firm tracks internally rather than what it publishes, and what it does when a search stalls.
How can I verify a matchmaker’s results?
Independently, you mostly cannot, and it is better to know that going in. What you can verify is process. Confirm how long the firm has operated and where. Read reviews on platforms it does not control alongside the ones it hosts. Ask who will handle your search and what the vetting confirms. Insist the term and the introduction count appear in writing.
Are client reviews better evidence than a success rate?
They are different evidence, and honest about being partial in a way a rate is not. A review describes one identifiable person’s experience and claims nothing about anyone else. A percentage claims to represent everyone while concealing the choices behind it. Reviews still carry a selection problem, and federal endorsement rules treat soliciting only from happy customers as potentially deceptive. Read them for specificity.
What outcome is reasonable to expect from working with a matchmaker?
A reasonable expectation is a series of introductions to people a matchmaker has met and screened, who are genuinely available and want the same kind of relationship you do, plus a process that adjusts after each introduction. That is what the fee buys. No service can promise that a particular person will be drawn to you, and a firm implying otherwise is more concerning than any number on its site.
Ready to talk?
If you want to try these questions on us before deciding anything, you can start with an application and we will take it from there.
No Comments