How To Get People To Reply To Your Messages In Online Dating, Part I

http://blog.okcupid.com/index.php/2009/09/03/how-to-get-people-to-reply-to-your-messages-in-online-dating-part-i/

by smokinn • 17 years ago

81 49 17 years ago

blog.okcupid.com

puredemo2• 17 years ago

This reads more like 'How to get lowest common denominator dates'.

If someone I write to can't be bothered to read ~700 words I'm fairly certain we aren't going to get along too well.

dotpavan• 17 years ago

to each his own, but what okcupid might be aiming at is something similar to comments-->leading to blog posts (if I strike a chord with your concise comment at HN, I would be more inclined to check your profile and follow links to say your blog etc and read more.. similarly for dating sites). this is like the trailer->movie effect which is an effect of short attention span and saves time.

GiraffeNecktie• 17 years ago

That may be true, but many people like to start off a little slower. If you write a lengthy essay about yourself, the other person might feel like it's too much too soon and they may feel pressured to reveal more about themselves than they're comfortable doing on first contact. It's got nothing to do with their reading ability.

puredemo2• 17 years ago

Sure, there is no need to tell your life story on first contact. But at least learning about interests is nice.

You can find out commonalities (as well as incompatibilities) sooner, and then shape the relationship along those lines.

mdakin• 17 years ago

It's true that "many people" may indeed like to start off slower but those data show clearly that for male initiated communication longer messages have increased probability of reply. The recommendation for short messages is intended to maximize replies per unit time not the reply of any given message.

I don't think there is a problem with either approach. Each strategy probably appeals to a different type of recipient. So the chosen strategy can be thought to partition the potential recipients into the different types. (If you send 'A' a long message and A replies A is of the type who likes long messages. 'A' very well might ignore all short messages. B might have the exact opposite preferences of A.)

At this level of abstraction there's not a wrong answer to the question. By that I mean that the details and content of a message whether long or short are probably more important than the length for any given case. But if you're someone who prefers to act in an open, earnest manner it's probably best to go with that and partition out people who are mistrustful, cynical or otherwise turned-off by open, earnest action and communication styles. An analogous statement holds for those who prefer to act more closeted, subtle and aloof.

philwelch3• 17 years ago

If someone writing to me can't be bothered to say something interesting in the first 100 words I'm fairly certain he won't get to anything interesting in the first 1000. You can be concise and intelligent.

scott_s• 17 years ago

I think you're interpreting the situation incorrectly. People probably aren't rejecting to reading a lot of text. They're probably using message length as a signal - people who share too much information with strangers put people off. It implies desperation at worst, and at best, an imbalance of interest.

roundsquare5• 17 years ago

If anything, the opposite of true. Long messages --> higher reply rate. Its the reply rate per message that is improved with shorter messages.

Edit: Of course this doesn't factor in the guilt mentioned by someone else, but with all these other variables I don't think this data is very useful as is.

dkarl3• 17 years ago

It's probably heresy here, but a lot of people still consider email a terrible way to get to know someone and online means in general to be extremely limited. An online dating site should be like Google: give me a bunch of options, and I'll make a quick low-bandwidth decision whether to invest in a face-to-face encounter. Engaging in an email conversation with someone isn't going to result in a better decision than making a snap judgment based on their picture and profile. I think it's more likely to result in a worse decision, actually. When I meet a potential date online, I just want to get face-to-face over drinks (coffee or alcohol) as quickly as possible so I can decide whether to have a real date with her.

For me, the only role of the initial contact is to say, "I would like to meet you. Would you like to meet me?" Of course, you also have to communicate that you're human and that you've actually read her profile, and you have to avoid communicating any desperation, narcissism, or other creepiness. So it's good to keep it short and make a little bit of conversation about something in her profile just to prove you read it.

theorique• 17 years ago

Definitely true.

Email is a good way to get to know someone on a technical level where you're kicking around programming problems or mathematical theorems. That's probably one reason why international open source collaboration works so well - people are (1) focused on the technical and (2) often share a common internationalized hacker/geek culture

On the other hand, to get to know someone in a romantic/sexual chemistry way, email is atrocious unless the people involved already share a lot of cultural reference points.

It's a lot like sending out a resume - you're not trying to pack everything about yourself into the written text, you're just trying to be interesting enough to get a phone call. With the phone call, you're trying to be interesting enough to get the low-risk face to face meeting. And so on...

imp• 17 years ago

So the summary is that a longer message has a higher chance for a response, but it's more efficient to just spew out a lot of shorter messages and spam everyone.

steveklabnik2• 17 years ago

This carries the implicit assumption that everyone you're messaging is equally important for you to get a date with.

philwelch3• 17 years ago

Overcommitting to any one person based upon an OkCupid profile without even having messaged them seems like a fallacy too. It's probably best to narrow down a reasonably sized bucket of people to message, without too many hopes or priorities about which one you end up with.

miloshh2• 17 years ago

It is assumed that longer messages take longer to compose. But that's only true if you're starting from scratch every time. You should instead write one good generic message and send it out to many people, only inserting one sentence to reflect on something specific to the particular profile.

msluyter2• 17 years ago

I think (smart) women can usually deduce that you've done that, and tend to react negatively.

msluyter2• 17 years ago

Yes, this was my first thought. If I was seriously stressed for time, then I'd agree with idea of optimizing for overall percentages. But if you really want to optimize the chances that a specific person will reply to you, then it seems as if longer is better. Still, I'd like to compare a short, witty, well-crafted, personally tailored message to a longer, more-generic one. (Not that that could be measured statistically...)

roundsquare5• 17 years ago

On that note, it'd be interesting to somehow mix this data in with (self reported) happiness of people who end up together. Not sure how... maybe happiness (1 - 10) vs length of original message.

dkarl3• 17 years ago

Long, rambling messages inspire feelings of pity and guilt. The longer the message, the more likely the recipient will feel obligated to send a polite response. Also, when a woman sends a polite negative response to a man, he's likely to respond with, "Thanks for responding, good luck in your search!" That shouldn't count as "starting a conversation," but according to this blog post it does.

Tichy• 17 years ago

More tips:

#1) Make sure the recipient is not a fake #2) Leave as much to the imagination as possible (actually that is probably why shorter messages work better)

erikb85• 17 years ago

I like to see that with actual data! Nice post!

I just want to add, if you write less words, every word counts more. So you must be more careful, too, if you write shorter messages. I wonder how the stats would look, if you take this in consideration, too.

alain940402• 17 years ago

Am I reading the graph wrong or is the author confused? To me, the first graph seems to say that the longer the message, the more likely you get a response.

But the text of the article says that the graph indicates that shorter messages are better.

Freaky• 17 years ago

> The graph clearly shows that in raw terms, it helps guys to write longer messages. But when we factor in the actual time it takes to type a given message, it becomes clear that in terms of time put in vs. conversations generated, shorter is actually better.

bkovitz2• 17 years ago

As I read the graph, it says, "Message length makes almost no difference, although very short messages are noticeably worse."

araneae• 17 years ago

I'm a little surprised that the word "zombie" is more of a turn-on for women than "literature."

jcl• 17 years ago

And I suddenly have much greater hope for humanity: "ur" is the biggest turn-off.

hughprime• 17 years ago

Yeah, women hate it when you try to start a conversation by talking about ancient Sumeria. (Cambodia, too, judging by the unpopularity of "wat".)

philh• 17 years ago

I imagine this is related to GSOH, where zombie references are automatically hilarious.

More surprising to me is that "pretty" is a win while "beautiful" is a loss. Is that coming on too strong?

kscaldef• 17 years ago

Keep in mind that "pretty" has multiple meanings.

roundsquare5• 17 years ago

Actually, I'd guess this is selection bias, people who like zombie movies searching for others who like zombie movies.

fnid• 17 years ago

I met my wife on okcupid. I think my message was over 700 words.

bkovitz2• 17 years ago

Does that page commit the fallacy of "making a lot out of what we can measure, because we can measure it" or what?

How about whether the message says something interesting? Something funny? Something that shows that you actually read the profile you're responding to?

You could call these things "subjective", but that would be fruitless and asinine.

angstrom• 17 years ago

Yes, like what's the actual success rate of finding a relationship worth pursuing from either approach.

Dilpil2• 17 years ago

Are we looking at the same data? Looks like a pretty flat line to me.

roundsquare5• 17 years ago

I dunno, it goes from 23 -> 35, ~50% increase in response.

korch• 17 years ago

This advice is all wrong. The real key to success is to show a photo of yourself in front of your uber-expensive sports car, which is parked in front of your grand mansion, and somehow mention your net worth is more than seven, preferably eight figures. The rest is merely the inevitable.

psyklic• 17 years ago

This assumes that message length is an important factor ... which I bet it isn't (excluding ultra-short and ultra-long messages). How about you correlate reply with more relevant things like canned message/attractive photo/cool profile/etc. -- or at least separate the length of reply vs response based on attractiveness or something.

You can get away with short messages as long as you show a genuine interest and not put out a canned message. And make some effort with a nice pic/interesting profile. This is the strategy that I use and it works out pretty well ...

Estragon• 17 years ago

My girlfriend messaged me on OKCupid because of something funny in my profile.

How To Get People To Reply To Your Messages In Online Dating, Part I

Related Stories