How many relationships REALLY start online? (Like, really.)

Thanks to an aggressive marketing campaign, industry leader now has everyone saying it: “1 in 5 relationships start online.” But is this for real? Like, really real?

My first encounter with the 1 in 5 stat was via a divorced software engineer on Cinco de Mayo 2010. Thanks to America’s love of alcoholidays, the restaurant patio where my friends and I met for drinks was packed. When a random guy asked if he could share the open side of the bench across from us, we said sure — the more the merrier.

Well, it turned out he didn’t want to sit down so much as seek refuge from an Internet date gone awry. While his date pounded tequila shots just around the corner (did I mention this guy was 47?), our group engaged in some casual banter about the situation. Meaning, my friends and I mercilessly cross-examined him on his choice of romantic strategy.

“Isn’t online dating just for crazy people?” I asked, in the classic kidding-but-not-really-kidding style.

“Everyone is doing it,” he countered, “These days, 1 in 5 relationships start online!”


If you’re old-ish like me (i.e. can remember life before smartphones and Facebook, when learning cursive was more important than words-per-minute type speed, and the clicks, squeaks, and dings of a dial-up connection), then you probably also remember when people considered Internet dating as um… kinda desperate. Basically, an approach used by “only the lonely” and those otherwise inept at personal relations.

1 in 5? Really?! Was a knee-jerk reaction, and a factoid I just had to fact-check.

When I got home that night, I started to investigate a few questions: Where did Match derive this statistic? How realistic is it? What do they define as a “relationship”?

To answer to the first question:

It turns out that the oft-quoted “1 in 5” advertisement is based on a survey conducted by Chadwick Martin Bailey.  Who is Chadwick Martin Bailey?  A for-hire research firm catering to corporate marketing needs.  While these type of industry commissioned statistics are dubious from a scientific credibility standpoint, for the most part they’re acceptable and legal in advertising.  The widely reiterated study results are available in a short 4-page PDF on’s website, and at first glance, seem semi-legitimate.

Thinking a little more on the topic, however, and attempting to answer the second question, things get shady:

For starters, the Chadwick Martin Bailey survey results came from an unspecified online Consumer Research Panel, aka a self-selected population of Internet users, aka the sample consisted of people who get paid to fill out surveys online.

Compared to the population as a whole, it’s not surprising that people who voluntarily complete Internet surveys for income might also go online for other things… like dating. It almost seems like going to an AA meeting and asking everyone in the room if they’ve ever been to a bar, a liquor store, or heck, even a wine-tasting.   We’d expect the answer to be “yes” with a pretty high frequency, right?

Additionally, the Chadwick Martin Bailey document showed the following numbers:

The “General Survey” portion used to support the 1 in 5 claim is based on (N = 2,525) panelists, (N=1,500) who were currently single and (N=1,025) who stated they were in a relationship.

Out of those (N=1,025), the reported incidence of relationships that started on an online dating site is in a lone two-by-three table on the fourth and final page.

The total amount of information provided includes the topic question — “Are currently in a relationship with someone met through an online dating site” — followed by responses: “N/A” for the 1,500 singles and “20%” for the 1,025 panelists in a relationship. Doing the math, out of 1,025 Consumer Research Panel coupled-up survey takers, 205 were currently in a relationship that started online.

Ok, so…205 people.

And, come to think of it, that’s 205 out of 2,525 overall — Chadwick Martin Bailey ONLY included data from people who said they are in a relationship, but conveniently didn’t bother to ask the rest of the sample if they have or are online dating. For the sake of argument, let’s assume that everyone in the sample searches for love on the net (if you’re online answering surveys for funsies, might as well keep a tab open for the The One).  Theoretically, that could be construed as 205 out of 2,525 people who go online find a relationship, which would be more like a 8.1% chance of finding a relationship online.

One in twelve relationships start online… Not nearly as catchy, is it?

The third question, what constitutes a “relationship,” remains largely undefined. We can only assume it was up for interpretation by the individual, and may range from 10 year happy couplings to week-long rendezvous. Without more information on their survey or sample, we’ll never know.

Overall, what we can know is this:

A skewed sample of respondents and 205 ambiguous answers were used to construct Match’s nationwide advertising campaign, and this data is the reason why millions of Americans believe that 1/5 of all couples met through an online matchmaking website.

Methinks there’s something wrong with this picture.


12 responses to “How many relationships REALLY start online? (Like, really.)

  1. Well, that’s stats for you. The first thing you learn studying PR, sociology or psychology is that you can very easily manipulate them, and why you shouldn’t.
    However, from my personal observation (which is a qualitative rather than quantitative method) I believe it’s quite a high percentage of recent relationships around me that were started like that, so who knows?!

    • Interesting you say that about recent relationships around you — I think it shows their marketing campaign worked! the whole 1 in 5 slogan was about making people believe online dating was “normal” and reducing the stigma so more people would join and be willing to talk about it…

  2. Pingback: Dating Via The Internet | Computers and Society·

  3. I’m currant puishing a book called Dating Decoded and I talk about and explain both online and off line dating. My research for online dating and site started in 2006 and I have also questioned if 1 in 5 is truly correct to reality, although its not, it’s all marketing in my opinion. However 98% of men and women who were interviewed for Dating Decoded were unsatified with their experience with online dating sites and that percent was from the research of 57 online dating sites. Furthermore, the research for any online networkings for some odd reason involved the FBI because of the electronic wall of the Internet, due to the safety and securities of those who use the Internet for dating.

    • @ Author101 – I’m very curious about your findings. For friends, I am a match maker and I’ve grown increasingly suspicious of the utility of on-line dating. If there’s any chance of getting in touch with you – that would be most welcomed. If nothing else but to cheer you on about your writing project.

  4. You were so right to question the legitimacy of this oft-repeated claim. There seems to be a general lack of criticality (though, ironically, a surfeit of criticism) and it’s nice to see this redressed! Question: How did you arrive at the figure 5,525 in the sentence, “Technically, that could be construed as 205 out of 5,525, which would be more like one in twelve relationships start online…” Is it simply a typo and the figure should be 2,525? One of the many odd aspects of the data is that a relationship that starts on-line would mean that both people in the relationship had to be on-line. Assuming a relationship consists of 2 people that’s 102.5 couples (205/2). What’s 1/2 a couple?!? A single person.

    • Good catch! Yes, that was a typo. And, for however much criticism or critique it receives, the interesting thing about Match’s claim is that it did exactly what it was intended to — Normalize online dating.

      As for the 205/2 question, I think we’re supposed to assume it’s 205 x 2, and that the respondents were not paired with other individuals in the sample. So, there would be 205 relationships with a total of 410 individuals involved (205 respondents plus the 205 sig o’s they met out in the wwwild world), not 102.5 relationships between 205 people. Although, if someone’s sig O doubled up — something that definitely happens online! — you could have 409 people involved in 205 relationships. That would make for a “half-hearted” equation ;)

  5. the thing that always got me about the ‘1 in 5 relationships start online,’ claim was that 4 out of 5 relationships don’t start online – so i never bothered with online dating, i’ve got a better chance of dating someone i met at work, in the park, on holiday etc.

  6. This article makes no sense. How about mentioning normal dates with no internet influence and their success rate?

    The important number is not the relationship number, but the dates, speed, and choice options that people have with online dating…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s