Circl.es Blog

When we mine our database for interesting relationships, we are typically looking for differences: like how different email domains have different levels of desirability, or how people in the Bay Area are more athletic than elsewhere. But sometimes, the lack of differences is just as telling. So when we did a data dive comparing straight and gay guys, what we found was suprising: despite all the stereotypes, there are very few differences between these two groups.

The data set consisted of over 4000 straight guys and over 1000 gay guys, and over 100,000 yes/no answers on Circl.es. We’re a small dating site (please tell your friends!), but this is enough data to draw conclusions.

Characteristics of Straight vs. Gay Guys

We looked everywhere for differences, but could find very few. The average height, 5′ 10″, was exactly the same for the two groups (less than a tenth of an inch of difference). The average age was 29. There were no differences in education or race.

When looking at body types of these two groups, differences were minimal, with gay guys slightly more likely to be “slim.”

Spiritually, gay guys are slightly more likely to be “agnostic” or “athiest,” but 15.3% of them still characterise themselves as “Christian” versus 19.6% for straight males. Not a huge difference.

One significant difference was politics. Gay guys are almost twice as likely to be Liberal and less likely to be apathetic – reflecting the fight for gay rights that continues to play out on the political scene.

Everyone Likes Family Guy

We also looked at the 20 most popular “Likes” for straight and gay guys on Circl.es. Shockingly, 60% of the Likes were common between the groups.

Gay guys and straight guys tend to like the same shows (Family Guy, South Park, The Office) and Barack Obama. Gay guys also really like Lady Gaga and Glee…. straight guys, not so much. But the commonalities between the two groups far outweigh the differences.

Who is Choosier?

Circl.es works by allowing users to click “yes” or “no” to people and matching up the people who have mutual interest (and blocking everyone else). Surprisingly, we found that gay guys are more selective, saying yes to almost 10% fewer people than straight guys. The stereotype of gay guys being “easier” or more open to meeting people is not supported on Circl.es

Another stereotype among gay guys is that they are image obsessed. So we decided to look at the yes/no data only for people who are fit, athletic, or slim – to see if gay guys are more likely to say yes to this group. They were, but only by 4% more – a small number, and exactly the same percentage as straight guys… eg, gay males were no more likely to click “yes” to fit, athletic, or slim people than straight males.

The Bottom Line: The Differences are Minimal

When it comes down to it, the differences are outweighed by the similarities. Ultimately, the only big difference between straight guys and gay guys on Circl.es is that one group likes women and the other likes men… and Lady Gaga.

Most of the action on Circl.es happens in San Francisco – it’s by far our most active city. There are many stereotypes about San Francisco, so today we decided to take a deep dive into our database and see if we can find any unique factors that sets this tribe of people apart. Our data set consists of 1500 San Franciscans, and 5500 people from everywhere else.

We started by looking at basic factors – age, height, sexuality – and we found no differences between the Bay Area and “Beyond.” The average age on Circl.es is around 29, the average male is 5′ 10″, and the average female is 5′ 5″. About 20% of users are gay.

The differences start to show up when we look at body types. San Franciscans are an athletic bunch, whereas “Beyond” reports themselves most commonly as “average.” Maybe it’s all the quad-burning hill climbs around here.

Men are especially strapping in San Francisco, with a full 45% choosing “athletic” as their body type, 10% more than non-San Franciscans.

Women in San Francisco are much more likely to be “slim,” “fit,” or “athletic,” than their counterparts elsewhere.

When it comes to religion, San Franciscans surprisingly aren’t that different from others. But we tend to be slightly less Christian.

A huge difference between the Bay Area and Beyond is the level of education San Franciscans achieve. 85% of San Franciscans on Circl.es have a college or professional degree. Outside of San Francisco, a third of users achieved only a high school diploma.

Finally, San Francisco is often stereotyped as being extremely Liberal. Unsurprisingly, this is confirmed in our data. Non-San Franciscans are much more likely to be apathetic.

Circl.es works by matching people who have mutual interest. So we have hundreds of thousands of “yes” and “no” clicks – when people both click yes to each other, that’s a match. We compared this data for San Franciscans versus everyone else, and we found that San Franciscan women are choosier than others – they click “yes” approximately 5% less – whereas men in San Francisco are slightly less choosy than non-San Franciscans.

Overall, according to our database, San Franciscans are athletic, well-educated, and liberal, with some very selective women. Despite what some may say, it seems like a pretty desirable dating pool.

In the last post, we looked at the Desirability Quotient for different email domains at Circl.es – the percent of people receiving yes clicks.  Today, we took a look at the flip side – the percentage of times people of each domain say yes.  We call this the Selectivity Quotient.  The data set is slightly adjusted from the last post, as we’ve removed a few outliers – such as people who say yes to everyone.

Here is what we found.  The y-axis is the percentage of “yes” clicks given by each domain.

On average, most people say yes to about a third of profiles they see.  Just as Gmail users were the most desirable group (minus “Other”), they were also the most selective.  Hotmail users were the least selective, saying yes to significantly more profiles than other domains.  We can make the data a bit more granular by breaking it down by gender:

Predictably,  men click yes to more profiles than women, and Hotmail and Yahoo men are most likely to click yes.  Hotmail women are the least selective, with “Other” women being most selective.  This mirrors our earlier post and suggests that more attractive people are more selective – which is good, otherwise they’d be overwhelmed with matches.

Why are some groups more selective than others?  Does something set Hotmail apart?  To investigate this, we decided to look at the characteristics of each group.  Surprisingly, we found no significant differences in age, height, or body type.  One area of difference we found was education.  Gmail users are much more likely to have a college education; Hotmail users are more likely than other groups to have a high school education.

Finally, we noticed a difference in politics.  Hotmail users are more apathetic than other groups, and less liberal.

Does this apathetic political skew translate into Hotmail users being more apathetic when choosing matches?  This analysis seems to support this conclusion.  Being apathetic isn’t a bad thing – in fact, on Circl.es, it could lead to more matches.

Circl.es is a simple site, but it generates a lot of data.  Specifically, it generated over 100,000 yes or no answers over the last month – people say yes when they’re interested in someone, and no if they’re not.  So, for each person, you have something we call the “Desirability Quotient” – the percentage of “yes” answers out of the total yes and nos that person receives.  On the flip side, each user has the “Selectivity Quotient” – the percentage of yes answers that person gives to other people.

In a series of blog posts, we’re going to start mining the data and seeing if we can draw any conclusions about desirability and selectivity of certain groups.  It’s not very useful data, but it is quite fascinating nevertheless.  The first grouping we’ll look at is email addresses.

Why email addresses? Because at Circl.es we’ve noticed a few things – the vast majority of spam reports and complaints come from a tiny minority of email domains (cough, Aol).  We also, like many tech-minded people, are fond of Gmail, and we were looking forward to blogging about how Gmail users are more attractive than other groups.  Unfortunately, that’s not what we found.  But don’t panic – we’ve still bested Yahoo.

Before we get started, some data.  Here are the major email domains at Circl.es:

Clearly, the site has a strong Gmail skew, which is to be expected – our core user is a young, college-educated, tech-savvy person in the Bay Area, and this person usually uses Gmail.  But we also have enough data about other domains to draw conclusions.

Here is the data.  The following chart shows the percentage of yes’s for each domain (the Desirability Quotient) in aggregate.

What does this show?  On average, about a third of people browsing a profile say “yes.”  Gmail does seem to be the most “desirable” of the major email domains, followed by Hotmail and then Yahoo.  ”Other” also does quite well – more on this later.  Here it is broken down by gender.

At least on the Circl.es platform, which skews younger, women get a lot more yes clicks than men (percentage-wise).  Yahoo men are the least desirable group, getting a full 5% fewer yes answers than Gmail men.  Yahoo women fare better, coming in ahead of Hotmail women.

The huge shocker is “other” and, specifically, women who have “other” email domains.  They get almost a full 10% more yes answers than average.  What explains this?  What are these email addresses?  We expected they would be work emails – @gap.com, @google.com – or personal sites.  Maybe these are career women who are too good for even Gmail accounts? But, this wasn’t the case.

The answer, after just a glance at the data, is clear, and it should have been obvious: .edu.  Most of the “other” emails from women are college email addresses – for women that are either in college or recent graduates who haven’t updated the email they use for their Facebook accounts.  It all makes sense – they skew younger, have more active Facebook profiles, and are well-educated.  The college coed wins again.

So Gmail users, fear not – you may not be at the top of the totem pole, but you’re doing well.  And if you want to be part of the most desirable group on Circl.es, hold onto that college email address for as long as possible.  Or go to graduate school.

In the next blog post, we’ll look at the Selectivity Quotient.  Are Gmail users pickier? Are men saying yes to everyone? Stay tuned – follow us on Twitter for updates.

Notes:

• Circl.es users can’t see each others’ email addresses (of course!) – only we can see this information in our database.
• Circl.es has an option for non-binary genders, but we don’t have enough data on these users to include them in this analysis.  We also didn’t have enough data to include AOL users.
• This is basic analysis, and not perfect.  Ideally this data would be adjusted for age and other factors – eg, certain email domains skew younger and this may be a confounding factor.
• Yahoo users – we actually think Yahoo Mail is kind of hipster, and you should stick with it.

Today we’re unveiling the “Share the Love” Campaign. Apparently this was once a Subaru campaign but we’re hoping they don’t sue us. The slogan is just too good.

The goal of the campaign is to build our userbase. We’re currently at 5,000 users with the vast majority of activity happening in San Francisco. If you’ve been using the site, you already know that this simply isn’t enough people to make a dating site successful. You don’t have enough density.

As I’ve mentioned before, it’s very hard to build a dating website. People don’t like to share dating sites, and when they find a match they deactivate.  Other sites address this problem by spending millions of dollars on advertising, but Circl.es is a free site so that’s not an option.  Some other sites show users who never signed up (or deactivated) in order to make you think there are more people on the site.  That’s not our style.  The only way for Circl.es to grow is through word of mouth.

Share the Love makes it a bit easier to share, and it creates clear benefits to sharing.  The biggest change is that we are now requiring users to share with one person to see matches.  You only have to tell one friend, and you see all future matches.  This is annoying, we know, but we also think it is pretty reasonable.  We don’t charge a dime – we simply ask that you help us make the site sustainable.  Once you can see matches, you can send them messages and everything else for free.

Screenshot from the Campaign.

We also offer things like advanced filters and trips for people who share with more than one person.  If you share with three people, we followup with you on whether your friends got matches – it’s anonymous, of course, but you’ll learn something like, “ten matches were made because you shared with five people.”  You don’t know who got the matches (that’s private), but it should make you feel good about sharing. If you share with ten people, we bump your snapshot to the front of peoples’ queues – this is a huge advantage if you want more matches.

This site is really a labor of love, so we hope you don’t mind sharing it with a few friends. It’s free, we haven’t been funded, and we don’t do anything sketchy.  A lot of people tell us to post matches on Facebook like Spotify songs or use our most attractive people to lure random people onto the site, but these strategies simply go against the big idea behind Circl.es: honest, simple, efficient matching, with no shenanigans.

If you’re a fan, help us make this site stick by telling your coolest friends to give it a try.

Circl.es is now requiring all users to upload at least three photos. This should enable people to make better choices about who they’re interested in – especially since some users have very restrictive privacy settings; you can barely see who they are.

Uploading files is a pain, so Circl.es makes it easy to just grab a few Facebook photos and use those. Just like past changes at Circl.es, it will take a few weeks for users to log in and select photos for their profiles.

As always, feel free to send feedback to justin@circl.es if you have an opinion on this change or spot anything buggy!

People who recently got matches noticed something new on Circl.es – you can now message matches directly from our site instead of from Facebook.  The reason we implemented this is because some users have their privacy settings set to not allow strangers to message them.

The Circl.es messaging system is incredibly simple.  When you send a message, the other person will be notified via email within two hours if they haven’t seen the message yet.  We added this delay to reduce spam and keep things chill… and to not be as annoying as OkCupid with endless emails.

The most important thing to know about sending messages on Circl.es is this: when matches expire in two weeks, your messages will be deleted.  This can be a very good thing if you and your match don’t hit it off, but it could also cause problems if you don’t exchange contact information.  So if you’re unable to get to the coffeeshop in two weeks, send a Facebook message or exchange emails before time runs out!

Today we are bringing Facebook Likes into the Circl.es equation.  When you log in, you will be asked to share this information.  Once we have your Likes, you can remove any that you don’t want to be shown on Circl.es, because we’ve all liked some strange stuff on Facebook – not all our Likes reflect our interests or personality.

But many of our Likes do say things about us – and the fact that Likes aren’t very serious makes them even better – they are just a collection of things that can help people understand us better.  More importantly, they can help us see whether or not we have mutual Likes, or things in common with our matches.

Likes can be anything – music, movies, places, things.  You can see if you both are Lady Gaga fans or both like that Vegan restaurant down the street.  You can spot fellow Virgin America fans or Apple fanboys and fangirls (fanpeople?).  We list mutual Likes first so you are sure to see things that you have in common.  If you match, you will instantly have something to talk about.

Bringing in Likes is going to improve the Circl.es experience.  Not only can you see more about potential matches, but you can click any Like and browse by it.  So if you Like “Snowboarding” and click it, it will show you other people nearby who are also into snowboarding.  If you Like a certain writer or blog, you can find fellow fans.

Likes appear as a bar of images under peoples' names.

Early on, it will be difficult to browse by Likes because it will take time for all users to log in and share this information.  Also, it takes a certain level of density, which Circl.es is still striving for (please tell your friends to join!).  But the pieces are in place for you to see how the feature will work going forward.

Simplicity is a fundamental goal at Circl.es, and using Likes enables us to maintain a simple, clean interface, while dramatically improving the experience.  I hope you Like it.

Dozens of people have requested a feature to express interest in friends on Circl.es (they are filtered out of the queue so you don’t see people you know).  My response was always, “just send them a message and ask them out,” but there is something nice about the Circl.es matching system, where there’s no risk in expressing interest in someone, because they only find out if it’s mutual.  And, as we all know, our Facebook friends aren’t always our “real” friends – many of us have Facebook friends we met once long ago and don’t even know.  Now there’s a way to say you’re interested in these people on Circl.es.

How it Works

You will never see Facebook friends in your queue, but now, on the homepage, you can enter your Facebook friends that you have an interest in.  If they are signed up for Circl.es, we will send them an email letting them know that “a friend” has expressed interest.  If they are also interested in you, you will get a match just like any other.

If they aren’t signed up for Circl.es, things get trickier.  Due to (prudent) Facebook privacy rules, there is no way for Circl.es to contact them and tell them they have a match.  If you give us an email, we will send them an anonymous message telling them someone on Circl.es is interested.  If not, we will encourage their friends who are already using Circl.es to get them to join – but this is hit or miss so it’s best to give us an email!

Go Forth and Fancy

Go ahead and add people – there’s no risk, only the potential for romance. And it’s fun! And if you don’t get a match, you can always revert back to the old fashioned way – asking them out directly.