The scorecard below is built for online matchmaking services: apps and web platforms where an algorithm or a coordinator selects from a pool. It works on a personal matchmaker too, though several dimensions score differently when a human is exercising judgement on your behalf.
Every matchmaking service describes itself with the same four words: curated, personalised, selective, high-calibre. Those words carry no information, because no operator has ever described its own pool as small or its screening as superficial.
A scorecard fixes that by forcing the same twelve questions onto every service you consider, including ones this series has never examined. The dimensions below come from what varies between operators rather than what they advertise. Each one resolves to something you can check or ask before paying. Score two services on the same twelve and the comparison stops being a matter of which website you found more convincing.
Why a scorecard beats a gut feeling
A sales page is built to produce a feeling, and it succeeds. You come away with an impression of quality assembled from typography, testimonial photography, and confident copy. None of that correlates with whether the service will introduce you to anyone worth meeting.
Selection is also where the returns are. Most of what makes a relationship work is decided by who you pick rather than by effort applied afterwards, and a good-fit match needs little maintenance while a bad-fit one absorbs enormous work that never resolves. A paid service is a tool for choosing who you meet, so checking the tool carefully is worth the hour it takes.
A scorecard replaces that impression with a list of answers. Some answers will be missing, and the gaps carry information too. A service that will not state its pool size in any form has told you something about its pool size.
The second benefit shows up when you compare two operators. Impressions cannot be compared, because you formed them at different times in different moods. Scores survive the comparison.
The twelve dimensions
Each dimension isolates one thing that varies between real services.
Candidate pool. How many active members, and how many fit your basic parameters? A large pool of the wrong people is a small pool.
Screening depth. What does the service verify, and how? Verification runs from documentary proof of income and employment down to a photo review, and operators use the same word for both ends of that range.
Transparency. Is pricing published before you apply? Are the terms readable and specific? Undisclosed pricing is common enough to be worth a dimension of its own.
Cost. Total cost to reach one meeting, including entry fees, membership, and per-introduction charges. Compare cost per meeting rather than headline price.
Convenience. How much coordination does the service remove? Scheduling, venue selection, and rebooking are the usual components.
Choice. Do you select from candidates, or does the service select for you? One profile at a time and a monthly batch are different products.
Privacy. Who sees your profile, what happens to your photos, and can you be discovered by people you know?
Safety. Identity verification, conduct policy, what happens after a bad meeting, and whether anyone answers when you report one.
Match methodology. What is the stated basis for a match, and does the explanation describe a mechanism or use adjectives? Claims about algorithms deserve the same scrutiny as claims about intuition.
Outcome evidence. What proportion of introductions produce a second meeting? Almost nobody publishes this. Ask anyway, because the response tells you whether they measure it.
Support. Who do you contact when something goes wrong, on what channel, and how fast do they answer?
Value. Given everything above, is the price defensible against the alternatives available to you specifically?
One caution on methodology. Screening measures the things anyone would recognise as impressive: income, job title, employer. Those are not the same as whether a particular person suits you. Someone can clear every filter the service applies and still be wrong for you, and that fit is what decides whether a second meeting happens.
The scorecard itself
This is how to choose a matchmaking service without relying on its sales page. Copy the table and fill it in for each one you are considering. Three levels per dimension, because finer gradations invent precision you do not have.
Dimension | Weak (1) | Adequate (2) | Strong (3) |
|---|---|---|---|
Candidate pool | No figure given | Range or approximate size stated | Size stated and filterable to your parameters |
Screening depth | Self-reported only | Photo or profile review | Documents verified, method stated |
Transparency | Price hidden until signup | Price published, terms vague | Price and terms published and specific |
Cost | Unclear total, open-ended | Known total, high per meeting | Known and competitive per meeting |
Convenience | You arrange everything | Partial coordination | Scheduling and venue handled |
Choice | No say in candidates | Accept or decline one at a time | Meaningful selection from several |
Privacy | No stated policy | Policy exists, generic | Specific controls, discoverability addressed |
Safety | No conduct or ID policy | ID checked | ID checked, conduct policy, reporting route |
Match methodology | Adjectives only | Stated criteria | Mechanism described and checkable |
Outcome evidence | None, and not measured | Anecdotes or testimonials | Measured rate, stated with method |
Support | Form only, no response time | Named channel | Named channel with stated response time |
Value | Price exceeds alternatives | Comparable to alternatives | Better than alternatives for your case |
Total runs from 12 to 36. The number matters less than where the low scores cluster.
How to score without fooling yourself
Three rules keep the exercise honest.
Score what you can verify, not what you were told. A claim on a sales page is a claim. If the service says it verifies income, the question is what document it asks for. No answer means score the dimension weak rather than adequate.
Score missing information as weak. The temptation is to leave a gap blank and mentally treat it as neutral. Absence of information about safety policy is information about safety policy.
Score before you talk to a salesperson, then again after. Salespeople run consultation calls to move you. Compare your two scores and you can see how much of the movement came from new facts rather than rapport.
Weighting: which dimensions matter for you
An unweighted total treats privacy and convenience as equally important. Which of them ranks higher depends on your situation rather than on the service.
Pick the three dimensions that matter most before you score anything. Someone in a small professional community may weight privacy above everything. A service scoring 30 overall but weak on privacy is the wrong service for them. Someone whose constraint is time will weight convenience and choice. Someone who has been burned will weight safety and outcome evidence.
Doing this after you have seen the scores lets you rationalise toward the service you already liked, which defeats the exercise. Decide the weights first and write them down.
One warning about the value dimension. Value depends on all eleven other answers, so scoring it early tends to colour the rest. Leave it until the other eleven are done.
What a low score does not mean
A low total does not mean a service is a scam. Small operators score badly on transparency and outcome evidence because they are small rather than secretive. A two-year-old company with no published success rate sits in the ordinary position of most new businesses.
The scorecard also cannot see whether you will like the person a service introduces you to. It measures the service, not the match. A service scoring 34 can introduce you to four people you have no interest in. A service scoring 20 can introduce you to one who matters. Selection quality at the individual level is not visible from outside.
The scorecard stops you paying for a service that fails on the dimension your situation depends on. That is the most common expensive mistake in this category. It narrows the field. Narrowing is the job; the choice stays yours.
Frequently Asked Questions
How do I choose a matchmaking service?
Score candidates on the same dimensions rather than comparing sales pages. Pick your three most important dimensions first. Then rate each service on pool, screening, transparency, cost, convenience, choice, privacy, safety, methodology, outcome evidence, support and value. Compare where the weak scores fall rather than the totals alone.
What questions should I ask before paying a matchmaker?
Ask the total cost to reach one meeting, what gets verified about other members, how many active members fit your parameters, and what share of introductions lead to a second date. The last one matters most, because the answer reveals whether they measure outcomes at all.
What is a normal success rate for matchmaking?
No reliable industry figure exists, and services that quote one rarely explain how it was measured. Treat any published rate as a claim requiring a method. The useful question is not the number but whether the operator tracks it.
Should I trust a service that will not publish its prices?
Undisclosed pricing is common and is not proof of bad faith. Some operators price per case. Treat it as one weak score rather than a disqualification, and ask for the total cost to reach one meeting before committing anything.
Is a matchmaker worth it compared to apps?
It depends which dimension your problem sits on. Matchmaking buys access, screening and coordination, so it is worth paying for when those are your constraints. It cannot improve what happens during the meeting, which is where many people's difficulty sits.
Conclusion
Deciding how to choose a matchmaking service comes down to refusing to evaluate it on the vocabulary it chose for itself. Twelve dimensions, three levels each, weighted by what your own situation requires, and scored before a consultation call rather than after one.
The output is a shortlist with the weak points visible. That is what lets you ask the two or three questions that would otherwise surface only after you had paid. Run the same twelve on the next operator and the comparison holds.
The next two articles apply this scorecard to two specific operators, in the same order, so the method stays separable from the verdicts. If access is your bottleneck and you would rather evaluate a room than a sales page, LoveLTR's Bangkok dating events publish format and crowd composition before booking.
