An ad for online dating site eHarmony has been banned by watchdogs because love can't be "scientifically proven."
The London Underground billboard suggested singles would have a much better a chance of finding lasting romance if they used the service.
But its claim to have a "scientifically proven matching system" was misleading, ruled the Advertising Standards Authority (ASA).
It upheld a complaint that it was not possible to hold scientific proof about a dating system.
The ASA said: "We told eHarmony to remove the claim 'scientifically proven matching system' and not to use similar claims with the same meaning, unless they had adequate evidence their website offered users a significantly greater chance of finding lasting love than what could be achieved if they didn't use the service."
'Comprehensive personality profile'
About 17 million Britons a year use dating sites. One in five relationships in the UK now begins online.
The website uses a matchmaking algorithm to pair people, based on an extensive list of personal preferences. The computer programme then spits out a list of suitable dates.
The relationship questionnaire, says the website, will "help us build a comprehensive personality profile of you and work out which of our members you will be deeply compatible with."
The Tube platform ad last July featured the headline "Step aside, fate. It's time science had a go at love".
It added: "Imagine being able to stack the odds of finding lasting love entirely in your favour. eHarmony's scientifically proven matching system decodes the mystery of compatibility and chemistry so you don't have to.
"Why leave the most important search of your life to chance?. Try something different today. Join eharmony.co.uk."
The ASA said the public was likely to appreciate eHarmony would not be able to guarantee love.
But they considered people would interpret the claim "scientifically proven matching system" to mean scientific studies had demonstrated it increased their chances.
They said: "We understood eHarmony used an algorithm that matched users with similar personality traits, values and interests to themselves based on questionnaire results."
The website submitted a published study which reported the findings of a questionnaire of 19,131 married respondents in the US, with the largest number of those meeting through online dating doing so through eHarmony.
But the figure was not higher than other general online and offline sources such as respondents who had met through online social networks, at work, through friends or in a bar or club.
The ASA said: "In any case, we considered the number of couples in one sample who had met through eHarmony would not in itself constitute proof that the website provided a greater chance of finding lasting love.
"We also noted that while the website had a lower percentage of marital break-ups than other dating websites, it had a higher percentage of marital break-ups compared to those who had met through email, online communities and through messages on blogs."
A second study found couples who had met through eHarmony had a significantly higher marital quality compared to couples who had met offline.
But this was not a random or representative sample, as they had been encouraged or incentivised by eHarmony to get in touch, and were more likely to report positively on their marital satisfaction.
The ASA said: "We further considered that both studies did not reveal anything about the percentage of the overall users of eHarmony who had found lasting love after using the website compared to other sources.
"Therefore, neither study provided insight into the likelihood of the website finding users lasting love compared to users who did not use the service.
"Because the evidence provided by eHarmony did not demonstrate that their matching system offered users a significantly greater chance of finding lasting love than what could be achieved if they didn't use the service, we concluded that the claim "scientifically proven matching system" was misleading."
'Based on scientific theories'
The website said the ad did not make any specific claims except that their matching system was scientific and could therefore provide an advantage in finding a compatible partner over a purely chance-based system or meeting.
They believed that consumers would interpret the ad to mean that the scientific approach eHarmony used could potentially work for them, but did not believe that consumers would interpret the ad to mean that it would guarantee they would find lasting love or make connections.
They explained their algorithm was based on data collected from more than 50,000 married couples in 23 different countries, which looked at their core personality traits and key values.
Their research resulted in statistical models which were associated with cut-off thresholds for scores that indicated a high probability of successful relationships if married.
They said the algorithm was based on scientific theories in the relationship literature of assortative mating.
But the ASA said the ad must not appear again in its current form.