Seventh Transparency Report

One of our key objectives is …

… to bring people together. To make sure that we can do this, it is important to prevent spammers and fakers from getting contact to real users. In this transparency report – the seventh already – we explain what changed in the anti-spam systems over the last quarter and which observations we made.

What has changed since the last anti-spam report?

The first period of 2018 has been a very exciting one for the anti-spam team at LOVOO. We worked on new machine learning models to detect more types of spam and scam and we also put a lot of effort into improving our existing models and heuristics.


In comparison to the last quarter the amount of spammers has remained stable. However, we have seen several small spikes as we added new features to our system.

Quarter Spam Users
Q3/2016 0.3%
Q4/2016 0.2%
Q1/2017 0.2%
Q2/2017 0.3%
Q3/2017 0.3%
Q4/2017 0.3%
Q1/2018 0.3%

The features we added were especially effective in finding and blocking less active fake profiles. While these profiles do not do any harm by themselves, they still annoy users as they may be visible in the nearby list, the radar, or the match game.

Spammer activity

Although there are only a few spam profiles, they are active above average and generate a fair amount of likes, i.e. positive votes in the Match.

Quarter Spam votes
Q3/2016 3%
Q4/2016 4%
Q1/2017 3%
Q2/2017 4%
Q3/2017 4%
Q4/2017 3%
Q1/2018 5%


The percentage of spam votes compared to the total votes is 4.81%. This is almost 1% more votes induced by spammers than in the previous quarter. While this is on the one hand  due to higher spam activity in the beginning of the year, it is also because our adjustments in AntiSpam allow us to identify more spammers than before.

Users meet fewer spammers

Spammers can be reported by users as well as identified automatically by the anti-spam system. Our goal is to block as many spammers as possible automatically, before the critical number of reports is reached or the support staff have to intervene manually.

Quarter AntiSpam
Q3/2016 76%
Q4/2016 90%
Q1/2017 85%
Q2/2017 89%
Q3/2017 87%
Q4/2017 85%
Q1/2018 87%

In this quarter we succeeded in blocking 87% of all spammers automatically using the anti-spam system. The remaining 13% spammers that got reported by users were also helpful. These reports help us to continuously improve our system. If we find a new type of spam or a completely new behavior of spammers, we include this information in the system which helps us to be able to recognize spammers sooner.  It also allows us to retroactively block profiles which showed the same characteristics in the past. Our main goal, however is to get the anti-spam system to find 100% of all spammers so that no user has to encounter them at all.


What happens to a report?

Quarter Reports Users Of which were no spammers
Q3/2016 1,051,560 819,710 82%
Q4/2016 1.100.966 859.483 81%
Q1/2017 1.317.714 990.074 80%
Q2/2017 1.145.683 883.431 79%
Q3/2017 1.385.517 1.057.697 77%
Q4/2017 1.292.563 1.122.601 65%
Q1/2018 1.088.465 997.987 63%

An often asked question is why a user is not blocked immediately after the first report. Very simple: users report other users for the most diverse reasons. Not all of these reasons justify blocking the reported user. Every day we receive a huge number of reports: In this quarter alone we got 1.088.465 reports that reported a staggering 997.987 of different user profiles. The positive thing about this number is: There are more than 200.000 fewer reports than last quarter, even though, the ratio of spammers/DAU remained constant and spammer activity has even increased this quarter. This means that we managed to detect spammers before they could annoy users to such an extent that reports were created. So while we had to supplement our anti-spam systems by relying on reports in the past, we are now much more efficient and less dependent on reports than in the previous quarters. Also the number of falsely reported profiles, i.e., profiles that turned out not to be spammers, continues to fall: only 63% of these profiles were no spammers. This helps us a lot as it means that we can trust these reports more and more. Especially in comparison to the number of falsely reported profiles from two years ago, the improvement is very obvious.

The anti-spam system is this fast on average:

We have to be very careful when blocking users automatically. On the one hand, we want to catch as many spammers as possible. On the other hand we need to be careful with our judgement to avoid blocking real users briefly exhibiting spam-like behavior. This can happen, for example, if Match is played very fast. The anti-spam system therefore waits until a user has exhibited negative behavior on several occasions before taking action.

Quarter Average duration
Q3/2016 2.2 h
Q4/2016 2.1 h
Q1/2017 1.1 h
Q2/2017 2.4 h
Q3/2017 2.7 h
Q4/2017 1.2 h
Q1/2018 0.7 h

On average in this quarter, we took 42 minutes from the first event executed, e.g. a vote in the Match, until the profile got blocked through the anti-spam system. This is significantly faster than in the previous quarter and, though also influenced by the spammer’s behavior, mainly a result of our continuous effort to train better machine learning models.

Are spammers more active on male or female accounts?

Quarter Sex Mean age
Q3/2016 ♀ 84% ♂ 16% ♀ 29 ♂ 29
Q4/2016 ♀ 70% ♂ 30% ♀ 31 ♂ 29
Q1/2017 ♀ 77% ♂ 23% ♀ 29 ♂ 29
Q2/2017 ♀ 80% ♂ 20% ♀ 29 ♂ 39
Q3/2017 ♀ 81% ♂ 19% ♀ 29 ♂ 40
Q4/2017 ♀ 81% ♂ 19% ♀ 30 ♂ 40
Q1/2018 ♀ 83% ♂ 17% ♀ 29 ♂ 38

As in the previous quarters, spammers had predominantly female profiles. Out of  all spam profiles 82.64% were female. On average these profiles were 29.27 years old. The male spam profiles were on average 10 years older. Why this discrepancy exists is a mystery.