20 Dating App Workers Reveal Things You May Not Know About The Industry

Congratulations! You’re now prepared to date. Perhaps, you’ll have a chance to meet someone with whom you can share the rest of your life. So you decided to sign up for online dating apps hoping to find someone who matches you, your personality, your ways of life, and your beliefs.

#1

Worked for Grindr a few years ago and any profile pic using that puppy tongue/ears Snapchat filter was an underage boy 98% of the time.

Source: bisexualfingerguns

#2

Image source: HueJass84

Most dating sites and apps are owned by one company The Match Group. They have a near monopoly. I think bumble is one of the few not owned by them.

#3

I wasn’t an employee but I was contracted by Match to run beer tasting events for them back when I did stand up and worked in craft beer. They split the groups for each event into age groups: 20-30, 30-40, 50+

The 20-30 group tended to be pretty chill for the most part, aside from there always being one super awkward dude who never spoke with any women and instead spent all the social portions trying to talk about craft beer with me. I’d always try to gently nudge them toward the single women and try to introduce them but they would usually either keep following me around or retreat to a corner and look at their phone.

The 30-40 group was an absolute nightmare. Regardless of gender everyone had this look of frenzied desperation in their eyes. I don’t know if it was about biological clocks or what but no one cared about the class itself and would just latch onto the first person they met regardless of chemistry (which there usually wasn’t).

The 50+ group was my absolute favorite. Everyone was chill, there was zero tension, they all just showed up, effortlessly made at least ten new friends and crushed craft beer for a couple hours.

Source: 420petkitties

#4

I worked for an online dating site 10 years ago on the IT side, here is a few things I remember:

– Most of the female users were fake. We would import thousands of fake profiles all the time to prop up the numbers and let the men think there were all of these women on the platform.
– Customer complaints were fun. The staff in that department were insane because they had seen and heard everything. The only way you got a refund is if you figured out all of the women were fake.
– We paid a local company to produce some “content” for the fake women. I had to move the video editors to a separate part of the office to edit videos because it was too distracting.
– We did a video of the week at our weekly meetings where videos that were complained about were shown to staff as a morale booster. It was pretty funny but sad.
– We weren’t allowed to use the dating site if we were employees. A programmer got fired because he was contacting women directly because he could look them up directly.

Oddly enough I met my girlfriend at the time on an online dating site, but through a competitor.

Source: matt95110

#5

Image source: Sighne, Tofros.com

A couple met on the dating app I worked on.
Unfortunately, the man passed away and the lady returned to the app where they met for remembrance.

One day, a bug in the system made some profile likes to be sent again after months and she received one from her deceased boyfriend.

Her bug report was heartbreaking.

#6

Ok so I didn’t work with a dating company per se… But I helped software engineers optimize their profiles.

Men get VERY FEW matches, regardless of how good their profile is.

Women get A LOT of matches, but most of those matches are useless.

—

Edit: this has gotten a lot of attention, so I wanted to share my basic advice.

1) Don’t try to appeal mildly to EVERYONE. Appeal strongly to a small subset of people. Emphasize who you are.

2) Show don’t tell — what makes a person want to date you? Will you impress them with your volunteer work? Will you bring them fun places? Will you make them laugh?

3) Get good photos. If you have to, get a friend with a good camera to take photos of you multiple times over a day with several changes of clothes.

4) be brutally honest. Do you need to go to the dentist because your teeth are gray? Go. Do you need to get a haircut? Go somewhere that charges $50 a haircut and tell them to do what they want (if male). Do your clothes fit? Ask a fashionable friend. Remember: people are judging you on your appearance as much as you are judging them. They can’t see you’re kind of funny or interesting. They can see if you’re well groomed and making an effort.

Any more advice and I charge $50 USD an hour ;)

Source: Katamende

#7

Image source: ShiversMTL, freestocks

Pick one! This was back in 2010, but we had bots acting as women sending messages to men, we spammed people on all social media sites and via text with alluring messages to get them to signup, we stored everyone’s password in plaintext and used it to login to their own email (about 50% of the passwords were the same password they’d use for their email) and target everyone on their contact list. We spammed so much, we spun off a second business just to handle all of the captchas, and that’s the only part of the business that still exists today (deathbycaptcha).

#8

Image source: Marzana1900, Kelli McClintock

I tested the communication feature for a dating company that shall remain nameless. Nothing unusual in itself. The message exchange function needed to be flawless (glitches ironed out etc.) Except in this case, the requirement was to establish a relationship with a user (usually 3 on the go simultaneously) and keep it going for about 2 months. Fake identity was used of course.

Months! I researched the person, likes, dislikes, interestes and so on. A whole fake relationship was build. Then, after the test run was over, ghost them, delete my profile and move on to the next subject. It was unnerving. Testing in this case is usually hit and run, but to latch on to a user and deceive them at length was just too creepy.

Couldn’t figure out the reasoning behind that. Put me off dating sites for good.

#9

This is my favorite bit from my time working at PeopleMedia, which is part of Match several years back as a software engineer.

One day while deep in the depths of code related to our spam filters (I forget what I was in there for) I stumbled across a curious code statement. It was like

“if (userId == xxxxx) return;”

Which is a very curious thing to see. It basically said that if the user was a certain person, don’t filter them as a spam profile.

Immediately I laughed and was like “uh… What in the actual &@$# is this?”

A few veterans explained that there was a guy who sent out messages at such high volume that he would constantly trigger the spam filters (which if you’ve ever been on a dating site is actually a kind of hard thing to do). And that he’d called up to complain multiple times, and they’d investigated his messaging and that not only was he messaging this many people, but that he wasn’t even copy/pasting his messages. Dude just really liked to reach out to women. Nothing inappropriate, just a true volume shooter.

So eventually someone just broke down and added a back door for him specifically to the spam filter.

It blew my mind that a single user had their own piece of code specifically to make their profile work in our code. Think millions of users across multiple sites, and this guy specifically had a little piece of it all to his own. That an engineer had taken time to write, and QA to test, and Devops to deploy. That alone had to easily offset any amount of money he’d paid over the years. But there it was.

Source: tex23bm

#10

Image source: SupermanistheDR, Priscilla Du Preez

We used to create fake accounts and chat with users. It was everything from someone having a premium account that wasn’t getting responses to bored employees.

#11

Image source: jamesinc, Anna Shvets

I ran operations for an online dating company (notably not affiliated with Match). From database analytics I can tell you a few things. Men initiate contact around 80% of the time in straight matchmaking, and if you are a woman looking to date other women and you simply initiate contact with another woman you have a good chance of success simply because it’s very very very common for women to match but then neither initiates contact. IIRC we were able to determine that it takes on average about 3 dates before sex happens (I don’t recall how we worked that out, I’m not a data analyst, but presumably it was some keyword based algorithm looking at chat messages).

We got so many requests for information from the police that we had an informal system with them, to save them from wasting time getting warrants for information about people who we didn’t have data on, they would ask about a particular name/email/whatever other identifier and we would just say yes we have data about them or no we don’t, and if we did they’d then go get the warrant to get a copy of it.

The other thing I can tell you from our analytics, that really shouldn’t be at all surprising, is to **get some decent profile photos.** Go get your talented friend or just hire a photographer to take some really nicely-lit well-composed photos of yourself and watch your match rate soar.

#12

I worked for Successful Singles in 2001. It was a dating agency. We cold called customers to get them to spend $3000 to be “professionally matched”. We would get some $$ if they showed up; we got more if they spent $ on the service. They claimed to have a “highly technological matching computer”. This was actually two high school girls in a room with two filing cabinets, one labeled male, one female. All leads came from a fake profile on match that said “send me your phone number so we can talk”.

I personally shut the place down after I was fired unlawfully. Called Fox 25 news undercover and Mike Bodet came out with a camera in a purse. Place closed two days after the report.

Source: KnobDingler

#13

I don’t know if it’s changed but (with a classic not me but) my roommate used to work at one of the big dating apps and one of the issues they had was that their algorithm changed at one point to more emphatically enforce dating “pools” where people who got more right swipes would only see profiles of people who get more right swipes etc. With the idea being that it would put people in similar “tiers” to actually match.

One big issue they were having was … well racial “preferences” or sexual racism being pretty amplified as a result. Black women and Asian men especially were being overwhelmingly shuffled down the algorithm because there are a lot of people who will basically automatically swipe left on them as soon as they see they’re a Black woman or an Asian man, even if they were hot as hell.

Also apparently, the issue was less severe among women seeking women but even more extreme (to an insane factor) among men seeking men.

Source: PhiloPhocion

#14

Image source: TheCharlienator, V T

I moderated a lesbian dating site for a short while and about 70% of the users were male fetishist, who would DM these women, thinking they would change their sexuality to do weird kink things with them. I don’t kink shame, these things were legitimately really weird, a lot of those DMs were straight up creepy.

#15

Image source: visualisewhirledpeas, NordWood Themes

My ex bf worked for the Yahoo Italy dating site back in the earlyish 2000s. His job was to pretend to be a woman, and message male customers just as their accounts were going to expire. This would encourage them to pay to renew their subscriptions. Once they renewed, he would ghost them.

He only lasted for a few months due to how unethical it was.

#16

Image source: anon, cottonbro

A dude with over 2000 right swipes and no matches.

#17

I used to work at Bumble, although this was about 4-5 years ago. Globally, about 90% of the users are men, so there is a huge male to female disparity, although it’s not that bad on a per country basis (for some countries).

The most depressing stat though was the histogram of word count in messages. Something like 91% of opening messages were just one word “hey”, and ~85% of conversations were just one exchange long (“hey” -> no reply ever).

Looking at human, digital mating habits splayed out in data science form was really depressing.

Some people have asked me a lot of the same questions so am providing answers here:

1) When I worked there, we were NOT allowed to read the content of chats, only gather metadata about them (word count, number of exchanges), but we could not build models which analysed the content of chats (this chat was about food, this one was about holidays, this one was sexual). This was due to stringent GDPR draft rules/TOS/privacy rules at the time, and Bumble took user privacy very seriously, so chats were never read or analysed for content, not even by automated models. However this was 4-5 years ago, and they may have amended their TOS since then to allow it, or they might be analysing content for non-GDPR countries (USA). Therefore, as others have pointed out, we don’t know for sure that the word used most often was actually “hey”, it could have been “hello” or “howdy”. In the office, we always assumed it was “hey” due to our own experiences on the app. But we did know with certainty that ~91% of first messages used only a single word, so we guessed it was “hey”.

2) The countries which had better male-female ratios (which I can remember) were the Nordic ones, Sweden and Norway were close to 50/50, and for a time, one of them even had more women on the app than men. Not sure how it is now.

3) The GDPR rules were released in 2016 (even earlier draft rules for some industries such as banking/finance and telecoms), but were not actually enforced until 2018. This grace period of two years was so that firms could implement all of the massive legal and technical protocols for GDPR, and many compliance/tech departments in London did nothing but a mad dash sprint for 2-3 years straight to get GDPR compliant before the 2018 deadline. As such, Bumble was already GDPR compliant (at least in terms of user chats, don’t know about other aspects) well before 2018 (as I said, they took user privacy seriously, even before GDPR). We weren’t allowed to read chat messages due to a combination of user privacy TOS (specific to Bumble) and GDPR draft rules. I just wrote GDPR as a catch-all term for all of the various privacy rules in effect then (including TOS), but yes, I agree, the rules were not enforced yet. I apologise that I wasn’t specific, I wrote my reply quickly this morning not thinking it would explode like this, and so just blurted out GDPR without getting pedantic about which rules in which years. Apologies for any confusion.

4) All these numbers are pulled from my memory of what I saw 4-5 years ago, and thus are not canon law or academic paper worthy.

Source: trias10

#18

Image source: throwaway492130921, Jonas Leupe

I worked for a dating app for a few years in a role that was pretty high up where I was privy to almost all of the inner workings of the app. I won’t say which one, but I think my experience probably is applicable to other apps as well.

1. We had a murder on our platform. The top of the company got interviewed as witnesses. TBH there wasn’t really anything we did our could have done about it, but it is crazy to think about.
2. One of our members got scammed out of six figures, and there was nothing we could do about it either. She was older, and lonely, and the person used an attractive picture and kind words to play off of that. If you let them, people will find any way to scam and abuse those who are lonely. Some of our systems for detecting and removing scammers and spammers were far more advanced than our systems for actually creating matches. Also we found older women were actually the most likely to be scammed. You can make your own conclusions from that.
3. To that point, the algorithms are less sophisticated than you think. They mostly consist of educated guesses, and then trial and error to see what creates the most engagement. This engagement could be anything from returning to the app, to sending messages. The main goal of the algorithm is always to get you to pay, never to actually ensure you meet somebody in real life, as much as we tried to lie to ourselves that it was.
4. No dating professionals or psychological professionals were ever consulted when we were building our software, software that basically plays cupid and changes the courses of peoples live. I kept thinking it would be a good idea to have experts and scientists tell us what determines attraction and sets up a relationship for success, but nobody was ever interested in hearing that. Instead we made our own choices about how to build this thing.
5. I met hundreds of our users in person, and they were all pretty great people. Many of them were willing to come in and talk because they were struggling with actually finding people and matches on our app. It was sad that our software was failing them, some of the best, most lovely people, really struggled to find a partner.
6. Almost every dating app has a significantly larger percentage of men than women.
7. We toyed with doing a test of “blind dating” where you couldn’t see a users profile picture until after a match, but that failed really quickly. People truly are superficial.
8. Contrary to many users on this thread, we were a large dating app and we didn’t actually create any fake accounts. We were certainly proud of that. That being said, there are some that do it, and it’s relatively obvious when they do. The profile is usually a very attractive person, somebody who…. probably has no need for dating apps, and it’s usually shown very early in the queue, and the photos tend to be of “instagram influencer” level quality. That’s the biggest giveaway.

I can likely answer questions too provided they are general enough.

#19

Female dating app users tend to sign off for the day several hours earlier than male users, which results in men who login after ~10PM generally not encountering many logged in female users.

In order to keep these men feeling like there is genuine female activity on the site (and thus continuing to pay for memberships), dating apps can pay for entire armies of “ghosts.”

Ghost profiles use photos of real women, but are operated by men, typically young men in their late teens and early 20s living in France, Serbia, Ukraine, and Russia.

A single ghost employee can manage dozens of female profiles, communicating with hundreds of men, for days or months at a time.

The ghosts’ general goal is to keep the men hooked on the site and still paying, but there are several possible unexpected consequences of this activity, none of which have been studied as far as I know:

1. Getting genuine responses, even from fake users, may help with feelings of loneliness.

2. A lot of these men start messaging the women with fairly gross or sexual messages (it’s usually late at night after all, when the men are bored, lonely, and horny). However, while genuine female users might take offense and block these men, the ghosts will instead roll with it and respond to messages like “nice mouthwatering tits” with “haha thank youuu. how are you feeling today? did you have a good day?” By turning the conversation around and encouraging the men to talk about their feelings, these ghosts may inadvertently be saving actual women from having to do this emotional labor.

3. While there’s a splendid irony in a secret cabal of men being paid to covertly perform emotional labor for other men, it’s fully possible that these ghosts are actually setting millions of men up for rude awakenings when they try to use similar language in real life with actual women who aren’t paid to put up with their overly sexualized nonsense.

Source: yetorico

#20

Image source: dazecoop, Radowan Nakif Rehan

I worked as a software engineer for a dating site in the mid-2000s. Literally every single female profile was fake, they were “generated” profiles using arbitrary data and paid-for lewd photos from various sources.

The sites we ran targeted guys paying for the site. Meanwhile if you signed up as a female, the site would be free — simply because the ratio guys to real-girls was so huge.

The fake female profiles would also message newly signed-up guys profiles almost straight away, giving the guys a confidence boost. They’d often message straight back, then after a few messages (which purposely got more “heated”, over a few different profiles), the guys’ account would get a notification saying to pay to send more messages.

If the guys’ profile didn’t pay, the site would continue to send messages from different fake profiles, all of which were behind a pay-wall in order to see more photos or reply at all.

The refund policy was super short, so if & by the time the guys’ profile realised he was just chatting to AI profiles, he’d request a refund, but often denied due to refund policy being short that he “agreed to” at sign-up.

By the time a given “site” was commonly known as being a scam, we’d spin up another site, different name, logo & design and repeat the process. This went on for months/years with 100’s of sites under various names.

TL;DR it was toxic as hell, hated it. Glad I quit.