Are My Husband and I Compatible? We Asked Science.

“Spit in this tube,” I said to my husband on a recent Wednesday night.

Spitting may not be the most romantic gesture out there (though like anything else, there seems to be an associated fetish….trust me, don’t Google that) but we were doing it as a couple nonetheless.

He took the proffered tube, but hesitated.  “If we don’t get good results, does this mean you’re going to divorce me?”

“No,” I said.  “But I’m going to write about it in my blog.”

ICKitWhen Instant Chemistry’s CEO reached out to me to see if I wanted to give their “couple compatibility” test a try, I jumped at the chance.  Using your DNA sample and responses to a personality quiz, Instant Chemistry looks at various genetic components and provides you with a “compatibility score”.

One of the scores is based on the HLA genes, which play a role in defining your immune system.  Instant Chemistry’s website says “many other studies have confirmed that up to 40% of physical attraction can be determined through our genes alone”.  I first heard about the idea of differing immune systems predicting physical attraction in early 2014.   The theory goes, the more different someone’s immune system is to yours, the more attracted to them you will be.  How can you tell if someone’s immune system is different?  You like the way they smell.

Sparked by this research, sweaty t-shirt parties became a trend.  Singles would bring in t-shirts they’d worn, which would be put in bags and marked with a number.  After rating the smell of all those t-shirts, they would be matched by mutual smell appreciation.

Ready to spit

A couple of weeks after I’d mailed in our spit tubes, I got an email saying our results had arrived.  I eagerly opened the attached report, and scanned the document for our compatibility score:  70%.

This score comes from three components:  bio compatibility, neuro compatibility, and psychological compatibility.  Each one of those is weighted differently.

Let’s look at “bio” compatibility first.  Our score was 76% (most long-term couples who have tried the kit have an average between 75% and 84%).

This is the one associated with those sweaty t-shirt studies:  it uses the HLA (immune) differences to calculate this score.  The higher the score, supposedly the higher the physical attraction.

Next is “neuro”.  Our score was 81%.  This one looks at serotonin, dopamine, oxytocin, and COMT.  The serotonin transporter gene “moderates the association between negative and positive emotional behavior”.  I have the “short” version, which allegedly means I have a strong response to emotional changes.  (For that, I’ll say “duh”.) My dopamine receptor labels me as a “builder”.  That makes me sensible with “good judgement”.  And not very impulsive.  Chris, on the other hand, was labeled an “explorer”:  “adventurous” and a risk-taker.

Finally, psych:  our score for this came out at 50%.  A strong component of this seems to be how social you are.  Although I’m an introvert, I am more social by nature than Chris; however, he’s also a very willing social partner.  If I suggest it, he’s up for it.  I think the questions may have overemphasized this difference.

Strangely enough, I spent very little time contemplating the “compatibility” piece of the scoring.  The explanations made enough sense to me that I didn’t really dwell on it.

But I became obsessed with one aspect that had nothing to do with our compatibility:  COMT.  You may have heard of this one before; there has been some recent media coverage of the “warrior” v “worrier” gene.  According to Instant Chemistry’s testing, I’m the one with the “warrior” and Chris the “worrier”.

I was so shocked by this I asked them to verify the tests to make sure we did not get our results mixed up, which they were kind enough to do (by checking the sex chromosome).

Obviously your genes are not the complete explanation for who you are, but I still find it incredibly hard to believe.  I always have a “top 10 worry list” in my head, while Chris rarely worries about anything. Instant Chemistry’s CEO, Ron Gonzalez, did comment further on Chris’s results:

When someone carries the worrier version of the COMT gene, there are higher levels of dopamine in the brain.  This is partly what causes the increase in anxiety as there are higher levels of dopamine acting on the brain.  However, he also carries the 7R+ version of DRD4.  This version of DRD4 is actually less sensitive to dopamine.  This would mean that even though there are higher levels of dopamine, characteristic of a worrier, the excess dopamine can’t affect the brain as much because he carries DRD4 7R+ which is less sensitive to dopamine.  His DRD4 gene is likely acting to partially cancel out the effect of the worrier gene.

So what do I make of all this?  I found it a very interesting exercise, but I’m not sure how much it’s going to change our behavior or our relationship.  I would have liked more concrete examples of how to apply the results to our day-to-day life.

I am still a fan of using the HLA component for dating:  not because I think it will predict whether you will be a great couple, but it does seem to have some potential for predicting physical attraction.  If you narrow down a number of people to go on first dates with, wouldn’t it be nice to know you chose someone you were likely to be more attracted to?

All in all, it was an interesting experience and I’m glad I took part.  It did cause me to reexamine parts of our personalities I’d always taken for granted and look at them in a new way.  What if I’m not as much of a worrier as I’ve always thought?  What if Chris reacts more negatively to stress than I’ve always assumed?

I love thinking about relationships and how they work.  Applying DNA analysis to this is an idea still in its infancy, but who knows, in 20 years maybe it’ll be as standard as a profile picture.  In addition to listing long walks on the beach under “likes”, you can specify your preference for a partner with a 7R+ dopamine receptor gene and the “Met” version of the catechol-O-methyl transferase enzyme.   Might make as much sense as dating someone because you both love Game of Thrones.

It’s Not About the App: Why You Can’t Get a 2nd Date

ShoppingCartDating apps are prolific.  New ones are emerging all the time.  There’s even a hugely popular podcast that follows the story of startup Dating Ring as the founders try to launch their app.

You would think that the only problems left were technical.  If someone can just figure out the right formula, the right user experience, then the problem will be solved.  Apps have all sorts of hooks:  go out on a date tonight, let women make the first move, take a 200+ personality questionnaire, create your own set of questions, and choose potential dates by their voice alone.

Yet so many people are still searching for love.

On Bravo’s TV show the Millionaire Matchmaker, Patti Stanger tries to find love for rich people who haven’t met their mate.  Over and over, her clients say they simply “haven’t met the right person”.  If only Patti could find them that person, they’d be ready to settle down in blissful true love.

But despite coaching by Patti on how to choose someone more compatible, how to conduct themselves on their first date, how to take it to the next level—most of the clients on the show don’t find “the one”.  They continue to choose people that initially attract them, knowing full well they won’t be compatible.  They interrogate their dates as if they’re interviewing for a job in the CIA.  They dismiss them for superficial reasons, such as not being tall/fit/successful/young enough.

The more I learn about what makes relationships successful, the more I circle back to the same conclusion:  it’s not that you haven’t met the right person.  It’s that everyone needs a class in How to Be a Good Date.

Apps focus on finding people.  Some focus on giving you lots of matches as quickly as possible (Tinder), and some focus on fewer, but supposedly higher-quality matches, much more slowly (eHarmony).

The problem is, we think the matter can be solved by this shopping approach.  With Tinder, even if you find someone you like, there is always the option of someone even better just a swipe away.  This is human nature.  Even in 1965, when the first computer-based matchmaking system matched two people successfully, they still entered their names during the second round, figuring since they’d done so well the first time around, they would do even better the second.  (They’ve now been married for 46 years, by the way.)

With sites like eHarmony, you’re under the illusion you can have a checklist of your perfect mate; it’s just a matter of finding the one who fills all the boxes.  If you’re not madly in love after date one, there are plenty of other fish in the sea.

Having too many choices is not always a good thing.  It can lead to indecision, waffling, and delay.  In addition, research increasingly shows it’s not about how compatible you are up front.  It’s not about having the exact same views on finances or travel, or about having the same Meyers-Brigg score.  A successful relationship is about how you treat each other, and how you treat the relationship.  Decades of research by John Gottman and his team says that the two most important factors in a successful relationship are kindness and generosity.

All this has got me thinking about how to really change the model.  Apps today are focused on finding people for you to date.  I’m not knocking that; it’s crucial to be able to expand your dating pool, especially once you’re out of school.  But then they just leave you hanging.  We need apps that help YOU become a better dater.  To coach you on how to be a good first date.  To give you the tools to know when someone would not be a good partner, despite being super hot and a great conversationalist.

I’m not saying we need to settle, or give up on finding someone we have great chemistry with.  But we need to get away  from the model of simply parading people in front of us, swiping left because they haven’t read Neuromancer (true story).  The idea of “there is always someone better out there, who will fit more of my list” is damaging.

Imagine an app that helps you analyze how your date went.  You can provide feedback on the person you went on a date with:  for example, did they talk about their ex too much?  Dominate the conversation?  Treat the waiter rudely?  When you shared the good news that you just got a promotion, did they say ‘That’s nice,’ and check their phone?

If you set up the date:  did you make it clear it’s a date?  Did you set up a time and place to meet well in advance, and arrive on time?  Did you ask your date questions without giving them the third degree?  Did you give someone a second chance that did not immediately make you hear wedding bells?

If you go on 20 dates and 15 people think you talk about work too much and 10 people think you dressed too casually:  wouldn’t you want to know?  Performance reviews at work are a standard.  Maybe it’s time for performance reviews in our relationships, as well.  (And they shouldn’t just stop once you’re married!)

Apps like The Grade allow you to rate users’ profiles, response rates, and message quality.  Mesh lets you set filters, such as how much vulgarity is ok with you, and the importance of grammar.  Bristlr calls out how many times someone has copied and pasted the exact same message they sent to you.  Bumble makes photos blurry until you consent to viewing them—and the photos are watermarked with the name and photo of the person who sent it..  These are awesome tools that can help weed out people who just aren’t ready for dating.

So a call to action to all the daters out there:  don’t make your “dealbreaker” list an impossible set of things to achieve.  Be a good date.  Give people a chance.  And above all, practice kindness and generosity.

Tinder Users More Likely to Tweet About Positive Things Than Negative

If I had to guess, I’d think people who were posting about Tinder would have a lot more negative things to say than positive.  Creepy messages, no-shows for dates, fake profiles… there are plenty of negative things out there.

But after digging into the data, I was surprised to find that, on average, the tweets about Tinder contain more positive words than negative:


I tagged words used in the tweets as either negative or positive, and increased the negative or positive score if the words had things like “very” in front of them.  The most negative words had a score of -1, and the most positive had a score of 1.

As you can see, most of the words fell into the slightly positive category, between 0.0 and 0.2.

I did the same thing for OkCupid and eHarmony tweets.  They had a generally similar trend, but were slightly more on the positive side than the Tinder tweets.

I wanted to know more.  What were the most common negative words in the tweets about Tinder, OkCupid, and eHarmony?  To find out, I created a word cloud.  The more common the word, the bigger it appears.


(A brief note about why some of the words look funny:  I used a technique called “stemming,” which groups similar words together by chopping off the end.  For example, “desper” includes desperate, desperation, etc.)

People are tweeting about some scary stuff!  Attack, scary, devil, panic, death.

Negative words like:  hate, weird, stupid, desperate.

Other things include:  garbage, fraud, freak, and drunk.

(By the way, the tagging set I used classifies the word ‘fun’ as positive and negative, presumably to include when people use it in the context of “making fun” of someone.)

What about the positive side?


Positive words include “funni” (for ‘funny’, ‘funnily’ etc), and of course, love.

Friend, success, amazing, hot, strong, caring, genuine–we can see what people are hoping to find when they tweet about online dating.

On a technical note:  I used the open-source tool KNIME to collect the tweets and do the analysis.  For more on how I did it, check out my blog post on the KNIME website.

A Dating Algorithm By Any Other Name

I recently gave a talk at a conference in Berlin about my findings analyzing Twitter data about online dating.  Afterwards a man came up to me and excitedly told me about an article he’d read in which someone had created an algorithm to “hack” dating sites, and landed himself a girlfriend. I was intrigued but skeptical, and when I got back to the States, I looked around and found it:


The author, a physics grad student, goes on to discuss the “algorithm” he created and successfully applied to dating sites that led to way more messages, more dates, and finally, a girlfriend.

You may have noticed I put algorithm in quotes.  The reason lies in the subtitle of his article:


Wait—that’s an algorithm?

Much has been made of the difficulties women have in online dating when they pass the ancient age of 30.  In this blog post from OkCupid, Christian Rudder does a fine job outlining the issue.  It’s not that there are no men of the same age who are single and looking to date; it’s that many men only want to date women younger than them.  Much younger:

As you can see, men tend to focus on the youngest women in their already skewed preference pool, and, what’s more, they spend a significant amount of energy pursuing women even younger than their stated minimum. No matter what he’s telling himself on his setting page, a 30 year-old man spends as much time messaging 18 and 19 year-olds as he does women his own age. On the other hand, women only a few years older are largely neglected.

From OkCupid OkTrends Blog
From OkCupid OkTrends Blog

The author of the Guardian article is in his 30s.  It would not surprise me if he’d set his search criteria for younger women.  He goes on to say (emphasis mine):

I realized that when I stormed out of the lab, I’d accidentally left off my search criteria during the auto-browse, and inadvertently discovered an incredibly powerful hack, a way to make the attention pyramid work for me. Over the course of 18 hours, my algorithm – logged in as me – had browsed thousands of active profiles, across all segments of women. These views didn’t pay attention to body type, race, or age, and mostly visited women that had just joined the site, or women that were high matches for me, many of them left wanting for attention by the usual online meat market.

As far as I can tell, he simply eliminated his search criteria and let the computer browse as many profiles as possible.  Since people can see who has browsed their profiles, many more women noticed he had looked at theirs.  Not surprisingly, this resulted in a lot more women messaging him (more in one night than he’d had in 3 months) and subsequently, more dates.

I don’t see an algorithm at all.  I see someone who expanded his pool of people he was interested in dating, which naturally led to more dates.  As Psychology Professor Eli Finkel has stated, online dating algorithms don’t work.  What DOES work is making the dating pool larger:

As a team of researchers… demonstrated, browsing profiles is virtually useless for discerning the sort of information that actually matters in a successful relationship. Curated text and a handful of pictures will never be able to tell you whether the first-date conversation will crackle or whether you’ll feel a desire to discover what makes this person tick.  The second faulty idea was that effective matchmaking algorithms could be based on information provided by individuals who were unaware of one another’s existence. One study…demonstrated that such information was highly ineffective at predicting initial attraction; another study found that such information was nearly useless in predicting satisfaction in long-term relationships. As almost a century of research on romantic relationships has taught us, predicting whether two people are romantically compatible requires the sort of information that comes to light only after they have actually met.

In additional, people are notoriously bad at determining what they want in a partner.  We can easily come up with a laundry list of “must have” traits, but when we finally end up with someone, he or she often doesn’t meet those criteria.  When dating sites allow you to narrow down your search to find your “perfect” candidate, they’re also doing users a disservice by eliminating people you would actually enjoy dating.

The man who created the bot ends his article on this note:

A pesky little voice pointed out that if I had gone outside once in a while instead of staying in my bed and coding maybe I would’ve run into her

Sounds like the best algorithm is to put down your smartphone, go outside, and meet people.

Are online or offline relationships more successful? The Jury’s Still Out

scientific_method_2Over the last year and a half, there have been two frequently-cited studies that compare relationship success between those that started offline vs. online.  The first is titled “Marital satisfaction and break-ups differ across on-line and off-line meeting venues” and was published in the June 2013 issue of the Proceedings of the National Academy of Sciences of the United States of America, and the other is titled “Is Online Better Than Offline for Meeting Partners?  Depends:  Are You Looking to Marry or Date?” and was published in the October 2014 issue of Cyberpsychology, Behavior, and Social Networking.

The results from these studies have been widely discussed in the Huffington Post, the Wall Street Journal, and the Washington Post, to name a few.

The conclusions of the two studies are contradictory.  The 2013 study proclaims that married people who met online have a higher marital satisfaction rate, as well as a lower divorce rate.  But the 2014 study says that couples who meet online are 3 times more likely to divorce.

What gives?

If you examine these studies a little more closely, it’s not so clear-cut.  Each one has flaws.

Problems With the 2013 Study

The 2013 study–the one that says online relationships are more successful–has come under fire because of the fact it was commissioned by eHarmony, and the lead author, John Cacioppo, was a paid advisor for them.  This doesn’t mean that the is study invalid, but it certainly calls it into question.

In addition, the results are not as noteworthy as the headlines make them out to be.  Marital breakup rates for those who met their spouses online (which could be anywhere, not just online dating sites) was 5.96%; for those who met offline, it was 7.67%.  That’s less than a 2% difference.  It is technically statistically significant, but as Professor Eli Finkel states, “Nobody’s surprised when a minuscule effect reaches statistical significance with a sample of 20,000 people, but it’s important that we don’t misunderstand ‘statistical significance’ to mean ‘practical significance.'”

Problems With the 2014 Study

As for the other study, which came out in the fall of 2014, graduate student Aditi Paul analyzed data collected from Stanford’s “How Couples Meet and Stay Together” to reach her conclusions.  Stanford collected data about how couples met, starting in 2009.  In 2010 and 2011, they followed up with couples to see if they were still together.

Sociologist Jessica Carbino says the methodology is “deeply flawed,” based on the fact the author uses only two demographic variables to explain relationship stability.

When I examined the study more closely, I was surprised by the small sample size.  The original data set, from 2009, had 2,923 people who identified as being in relationships:  of those, only 280 had met online (90 of those were married).  In 2010, in the first followup, none of the people who were married had broken up; in the second followup, conducted in 2011, 8% of the married people who had met online were divorced or separated, compared to 2% of those who’d met offline.  A compelling finding, but given there were only 90 people who fit the criteria of being married and meeting online, that means about 7 people were divorced.  I’d take that result with a giant grain of salt.

Another result is that after the first year, 32% of people in non-married relationships who met online had broken up, compared to 23% of those who had met offline.  However, after two years, there was no statistically significant difference in breakups between the two.

It is important to examine the differences between people who use online dating to find a long-term relationship and those who do not.  It’s quite possible that those who choose online dating are younger and less interested in a serious long-term relationship.  Although breaking up is not generally a desired outcome for most people, it is the natural result of wanting to be in a relationship, yet not being ready to settle down.

The results of the second study are being portrayed as though choosing to date online lessens your chance for a serious relationship.  But it does not take into account people who want to date a lot, and use online dating to do just that.  There are many possible reasons people might break up, but not all of them should be viewed as failures.

Don’t Give Up On Online Dating

Don’t get caught up in the hype the media is making of these studies.   They are a good start, but it’s too soon to make any major conclusions.  We still don’t know exactly why the two studies had different results.  We need more data.  In addition, the 2014 study looks at couples who were together in 2009–not very long ago in terms of human history, but a lifetime in terms of online dating.  Tinder was not available until August of 2012.  Mobile app dating had not taken off yet.  Perception of online dating was more negative.  Other factors include changing attitudes towards marriage: fewer people are marrying, people are marrying much later, and the divorce rate is declining.

Some estimates say 70% of couples will meet online by 2040.  In the not too-distant future, meeting your spouse somewhere other than online will seem strange.

Put a bunch of people together–in a school, a workplace, an online community, heck, even the grocery store–and relationships will be formed.  Breakups will happen.  Some marriages will flourish; others will crash.  Meeting people online is just one more place the complex world of relationships takes place.

eHarmony’s “Big Data” Talk: Keeping it Real for the Non-Techies

David Gevorkyan, a Principal Software Engineer at eHarmony, recently gave a talk discussing “how Hadoop helps [eHarmony] to process over a billion possible matches into several highly compatible matches for each of our users per day.”  Sounds pretty technical, right?

I watched the whole talk (53 minutes!)  and I’ve pulled out some pieces for the non-techies out there.  There were a lot of interesting tidbits about how eHarmony works.  You can see the talk, and the slides, on eHarmony’s engineering blog.

First off, I’m very pleased eHarmony put something out there that gives us a little bit more knowledge about how they work.  Transparency is a beautiful thing.  Also, thanks so much to David, who was kind enough to answer some of my questions about eHarmony and his talk.

Now, on to the good stuff!

Dr. Neil Clark Warren, founder of eHarmony, came up with a way to systematically match people, using “29 dimensions of compatibility”.  The exact 29 dimensions are not disclosed, but they include such things as humor, spirituality, sociability, and ambition.

Over 600,000 marriages have come from people meeting via eHarmony, or about 438 marriages per day (this accounts for about 5% of all new US marriages).  eHarmony currently has about 50 million registered users.

David mentioned a study conducted by Harris Interactive  for eHarmony that did an analysis on divorce rates, and for the 7-year period eHarmony has been operating, the divorce rate was about 4.8%.  (Statistics about current national divorce rates vary, but some recent research puts it at about a 40-50% chance during one’s lifetime.. so that’s looking at marriages much longer than 7 years.)

David says that what differentiates eHarmony from other matchmaking sites like and OkCupid is eHarmony’s “compatibility matching system,” which has three parts:

  1. Compatibility matching: compatibility based on the personality and psychological profiles
  2. Affinity matching: historical data from the last 15 years that uses machine learning models to predict different things such as probability of communication between users
  3. Match distribution: ensuring we deliver the right matches at the right time to as many people as possible throughout the entire network

Step 1:  Compatibility Matching

When you join eHarmony, you provide criteria such as preferences on distance, income, age range, religion, smoking and drinking preferences, and others.  After that, you fill in a comprehensive relationship questionnaire (150 questions!), which is targeted to extract personality and psychological profiles.  These questions provide eHarmony with information about personality, values, attributes, and beliefs.  eHarmony then uses the “29 dimensions of compatibility” to make the matches.

Sample eHarmony question

Based on a marital satisfaction survey of 5000 users, eHarmony took the most highly-satisfied couples and uses their compatibility scores to predict new matches.

When a new user joins eHarmony, it runs them through “complex mathematical equations”, which produces a score–if the score is above the threshold for the highly-satisfied couples from the survey, it considers them compatible.

David shared with me the link to one of eHarmony’s matching patents.

eHarmony matching patent

On a technical note, eHarmony uses a data storage system called Voldemort (developed by LinkedIn) to store its one-billion+ potential matches per day.

Step 2: Affinity Matching

Based on 15 years of historical data, the system will predict probability of communication between two users (among other things).  David says, ““Even though the users are compatible with each other, you might not always decide to give that user as a match.”

And why not?  Well, it may be that the user has specified he/she will only communicate with someone within a certain distance, or a certain age range.  So the system won’t try and match these people.  David told me there is some flexibility with this, but if a person has listed something has “very important” then eHarmony won’t give you a match that doesn’t meet your criteria.

He showed an interesting slide on how distance in miles affects the probability of communication:  most communication happens, not surprisingly, when users are nearer to each other.  However, at some point (over about 1000 miles) it doesn’t really matter any more–I guess long distance is long distance!

How distance affects probability of communicating

David says most communication happens when the man is taller by 4 to 8 inches–and that men are more eager to talk to women who are taller than them than women are to talk to men who are shorter than them.

Different words you use to describe yourself in your profile affect the probability of communication–that is, how likely you are to get a message from someone else.

For men, these words are likely to get more messages:  “perceptive, physically fit, passionate, intelligent, funny, optimistic”.  And for women:  “sweet, funny, ambitious, thoughtful, passionate”.

Each user has an average of about 1000 attributes, and altogether the users have answered about 4 billion questions.  eHarmony makes tens of millions of potential daily matches.  Now that’s a lot of data!

Technical note:  originally they were using Amazon Web Services, but one issue was that they could not predict when processing jobs (such as predicting matches) would finish.  Why does it matter?  They want to deliver potential matches “first thing in the morning”.

Step 3:  Match Distribution

eHarmony wants to make as many people on the system happy, so it tries to maximize communication between users.  This is done via machine learning to try and determine how many matches to send per day, what time of day, etc.

Finally, someone in the audience asked why certain people are rejected by eHarmony.  David said they do have machine learning algorithms in place that are a part of that, but did not give details.

My Ten Favorite Recent Articles on Love and Relationships

There are a ton of great (and not so great) articles out there about love, online dating, and the science of relationships.  I’d love to write posts on all of the interesting ones, but since that’s not feasible, I will instead share ten of my recent favorites.  Enjoy!

In no particular order…

Online Dating: Love in the Time of the Internet
A look at how online dating doesn’t let us use one of our most powerful tools:  intuition.

Matchmaker, Matchmaker, Make Me a Spreadsheet
Since OkCupid founder Christian Rudder’s new book, Dataclasym: Who We Are When We Think No One’s Looking just came out, he’s been doing a lot of interviews.  This one from Five Thirty Eight was very well-done.

Matching Algorithms, A Work in Progress
A thoughtful look from Online Dating Insider David Evans about the state of matching algorithms in online dating.  I’m definitely in agreement.

Tinder And Evolutionary Psychology
Liraz Margalit says Tinder gives us everything we need “to make an informed first impression about a potential long-term mate”.

Esther Perel: The secret to desire in a long-term relationship
A 20-minute TED talk on how long-term relationships affect desire.

For the first time, there are more single American adults than married ones, and here’s where they live
What affect will this have on marriage?  For more insights on what this means for single women, check out this NPR clip.

T is for Turning
Zach Brittle from the Gottman Institute Relationship blog talks about one of the things successful couples are better at than their non-successful counterparts:  turning towards their partner.

Infographic: The 10 Most Interesting Dating Studies of 2014
From the Science of Relationships blog, a round-up in infographics of interesting dating studies from 2014.

Couples on Different Sleep Schedules Can Expect Conflict—and Adapt
Most interesting quote from this Wall Street Journal article:  “when women reported higher relationship satisfaction, they were more likely to have been asleep at the same time as their partner the night before, almost down to the minute”.

How I Rebuilt Tinder And Discovered The Shameful Secret Of Attraction
An imperfect study, but it opens the door to some fascinating–and disturbing–possibilities of why we’re attracted to some people and not others.

Read anything interesting lately?  Send links my way at, or on Twitter @lovedatablog.

Common Problems in Online Dating–And How To Fix Them

ProblemsOnline dating is more popular than ever.  One third of all marriages these days came from online dating.  The stigma of meeting someone over the Internet has faded.   But it’s not all candlelit dinners and walks on the beach.  Online dating certainly helps people find potential dates, but it doesn’t always help people find good relationships.  There are a number of common problems out there, and there are ways the online dating industry can address them.

Problem:  Women Get Way Too Many Messages

On sites with unrestricted messaging (such as OkCupid and Plenty of Fish), women get way too many messages–sometimes 50-100 per day if the profile is particularly attractive.  That’s just too many to deal with.  This turns into a problem for men, as well–it’s much harder for a man’s message to get a response.

Some sites deal with this by only giving you a few matches per day (such as eHarmony) or only one (such as Coffee Meets Bagel).    This certainly helps reduce the onslaught of messages, but a lot of people still want to feel they have more options.

Solution:  Provide Feedback

One way a dating site could help is by noticing male users who have a poor ratio of messages sent to message replies.  The site could give them advice on the best way to craft their messages, such as this blog post from OkCupid’s OkTrends blog, and this TED talk by Evan Marc Katz.

Of course, lots of people will still ignore this advice, so in addition, dating sites could crowdsource this issue.  By allowing women to rate messages–with categories such as ‘overtly sexual’ or ‘generic’ (‘hey girl what’s up’)–the system could track this information.  If a user gets too many ratings in a particular category, he could get a warning/advice, and if he continues, the number of messages he’s allowed to send could be restricted.

Problem:  Men Send Creepy or Unsolicited Sexual Messages

Some men send messages that are sexually explicit to women they’ve never had contact with, or after only a few brief interactions.  This can be annoying at best, but can be very disturbing, and has driven women to quit online dating sites entirely.

Solution:  The Scarlet Letter

Dating sites should allow its users to easily (and anonymously) let the system know when they’re getting these type of unwelcome messages.  Users could check a box on the offending user’s profile labeling them as an “inappropriate message sender”.  If the user gets this feedback enough times, his profile would be labeled as such, so that women would see he’s prone to that type of behavior and avoid him.  If it continues, the number of messages he’s allowed to send could be restricted.

This feedback would be anonymous, of course.  In addition, if this happens after the messaging has gone off-site (e.g. you exchange phone numbers and he starts texting unwanted images), there should be a way to provide that feedback as well.

Problem:  The Current Algorithms Aren’t Working

Although places like eHarmony, and OkCupid claim to have algorithms that will help you find your match, there is no proof the algorithms actually work.  It’s true many relationships are formed by people meeting on these sites; what’s unproven is whether the site’s algorithms have anything to do with this.

Part of the problem is how difficult it is to predict compatibility between two individuals.  Having things in common (or not) is not an indicator that two people will find each other attractive.  The industry needs more couple-level data.

Solution:   Post-Date Feedback

There is e a LOT of data from online dating, but one place it’s really lacking is post-date feedback.  It’s hard to get people to provide that info, and there is no way for an online dating site to track user’s behavior once they move to offline interactions.

But this data is incredibly important.  Who did you go on a date with?  A second date?  Who did you find compatible, but without chemistry?  If the dating site could add this information to its algorithms, it would have a huge amount of potential to improve its matches.

If the dating site has “freemium” services, these can be used as incentives–the more dates you rate, the more perks you receive.  The site needs to be smart about its user experience when asking for the feedback.  It could notice if you’ve messaged with a particular user a lot, and ask if you’ve been on a date yet.  For sites like Coffee Meets Bagel, it could send a text the next day with a short survey (maybe two questions) to see how the date went.

It’s important to make the survey short and simple.  Anything longer than a few questions will become annoying very quickly.  Timing is important.  The best incentive to provide feedback, of course, is to show it’s working–over time, your matches should improve.  What could be more motivating than that?

No Need to Smell Sweaty T-Shirts Anymore.. Just Spit in a Tube

InstantChemistry uses DNA to determine compatibility

Back in February, I wrote about my idea of using pheromones to help predict chemistry between online daters.  Not surprisingly. it turns out there are several companies doing just that:  GenePartner, LoveGene, and InstantChemistry.  You order a kit, spit in a tube, and send your DNA off for analysis. (Better than my suggestion of taking a blood sample!)

The premise is simple:  couples with different immune systems produce healthier offspring, and thus we are naturally attracted to a person whose immune system (measured, in this case, by something called the MHC: major histocompatibility complex) is different than ours.

GenePartner got a lot of press back in 2009, but since then they’ve been pretty quiet.  LoveGene, based in the UK, has been around for a couple of years but seems to be getting off the ground just now–they’re getting more attention in the news. 

But InstantChemistry has been getting the most buzz.  In addition to measuring the immune system genes, they look at another one:  the serotonin transporter gene.  According to InstantChemistry:

When couples respond very strongly or differently to emotional situations, research has shown this creates conflict in relationships.  Over time, couples reported decreases in their relationship satisfaction if they had conflict in how they responded to emotional situations.

Because of this, they claim to be able to use DNA to “discover your relationship compatibility”.   You and your partner can both take the test ($215) and see if you match.  And if not, “we give you ways to potentially mitigate and improve your relationship.”  They are definitely upping the game with their offerings.

They’ve also partnered with some online dating sites, such as SingldOut.  SingldOut launched in July, so it’s too soon to see if it’s working for their daters, but Business Insider reports they’ve sent out more than 200 kits already. 

I love the idea of finding ways to predict chemistry between two people before they meet up in person.  It’s an important thing to know and you often can’t tell just by looking at someone’s profile or even talking on the phone.  I hope this is one more tool to make it easier for people to find love.

Sweaty t-shirt parties, by the way, continue.  (They are also called by the more appetizing “pheromone dating”.)  I enjoyed reading a first-person perspective on attending one of these parties by Katherine Templar Lewis.  When she finally got to meet the guy whose smelly t-shirt she’d picked (#112) she did in fact find him attractive.  But it only took a few minutes of conversation to discover, sadly–they didn’t have chemistry.


An Interview with Whim founder and CEO Eve Peters

whimLogoWhim founder and CEO Eve Peters graciously took the time to answer some questions for Love Data.  Whim is a new dating app that bills itself as “The ‘Hotel Tonight’ of Dating”.  They will send you on a date:  tonight!


Q: Given that there are over 3800 dating businesses out there right now, what made you decide to create something new in the dating space?

There may be 3800 dating businesses, but none of them is really solving the problem of meeting new people. While most apps have done a great job of mastering search and discovery of people, they’re really only solving 10% of the problem. 90% of the friction is in getting from the online connection to an offline relationship; Whim is uniquely focusing on providing a solution for that.

Q: In your CrowdFunder video ( you mention using 16 variables to help match potential dates. What sets Whim’s algorithm apart from other dating apps?

Most mobile apps will present you with any matches who fit your very basic criteria – like age range and location. We go further by employing the detailed level of matching you’d find on an app like Match (including race, education level, height, body type, interests, and more) while still providing the convenience of a mobile app.

Q: At the iDate conference in June, you had a great slide showing the typical time a person has to invest before getting to that first date. With the exception of Grindr, it seems to be at least 5 days to weeks before people usually meet up in person. Are you worried it will be more awkward for people on a Whim first date since they have not had any time to get to know a little bit about each other first?

I’m not concerned about that. First dates are a little silly feeling and awkward no matter what. The advantage of a Whim date is that it happens almost on the fly; there’s less time to get nervous, less build-up, and less of a need for it to be something great. It’s just a casual weeknight drink with someone you might enjoy.

Q: How many dates do you think the typical Whim user would want to go on? Do you think people may suffer from “first date fatigue”?

Our target is to get active users going on at least 2 Whim dates per week. They’re so easy to set up and so fun to go on – I think this goal is realistic. Most of the dating fatigue in apps comes from the extensive profile browsing and messaging back and forth that ensues, not from the actual face-to-face meeting. Of course, some people have more social energy than others; some Whim users might just do a couple per month, while others might date every other night like it’s their job!

Q: Can anyone with a Facebook account join Whim, or is there any kind of vetting process to help weed out less savory users?

Currently anyone with a FB account can join. However, if a user reports any suspicious or inappropriate behavior to us, we will immediately check it out and dismantle your profile if necessary.

Q: Whim has a “How was your date?” rating afterwards. How will this information be used to improve Whim’s algorithm or other features?

When enough data accumulates, we can use these ratings to predict what kinds of people you are most likely to enjoy on a date and serve you continually better matches.

Q: Whim’s model includes potential partnerships with bars and restaurants. You pay $5 if you accept a date for that night, and in turn the bar/restaurant offers a free drink or appetizer. I love this idea because it cuts down on people flaking, and has an actual revenue model for Whim–something a lot of dating apps lack. Do you have a lot of partnerships with bars/restaurants at this time?

We have ten partnerships established thus far, including Primitivo, Salt Air, Willie Jane, A Frame, and Circle Bar (all Los Angeles establishments).

Q: Congrats on getting a slot with 500 Startups! What does this mean for Whim’s next steps?

Thank you! As of July 14th the team has moved up to San Francisco. We’re grateful and excited to be part of 500 Startups; it means we’ll have the advantage of their mentorship, network, and support behind us as we move forward toward our native app launch (September ‘14) and beyond.