US Edition 
Post on Facebook - and get a tax bill
Banks, insurers and even the taxman are trawling Facebook and other sites looking for information which could determine how much they charge you
Send one wrong tweet or message and you could lose your reputation, friends and your job.
It's the phenomenon of "social shaming", in which an errant post or message can instantly be shared and seen by millions of people.
Its immediacy was demonstrated during a recent sexism row when Charlotte Proudman, 27 (pictured below), publicised fellow barrister Alexander Carter-Silk on social media for sending her a "misogynistic" message on LinkedIn.
Offended by Mr Carter-Silk's contact on the professional network, in which he complimented her "stunning" profile picture, she published its contents on her Twitter page.
But our tweets, posts and messages haven't just grabbed the attention of the public – big businesses, too, are watching our online behaviour.
Banks, insurers and Government bodies are trawling the internet for any information we give away on our social media accounts to price products and catch anyone who cheats the system.
The City regulator has set its sights on whether insurance companies in particular are using Facebook and Twitter to unfairly increase premiums, for example.
The Financial Conduct Authority (FCA) said it is investigating concerns over insurers using social media profiles to "risk-assess" applicants. It says it will report findings by January and investigate further if necessary.
But it doesn't stop there. The taxman and mortgage lenders are also going to great lengths to harvest our personal data, Telegraph Money can disclose.
Here, we detail exactly what organisations are doing now – and how they are about to become more vigilant.
1) Discovering tax dodgers
Tax dodgers who boast about their luxury holidays and spending sprees on social media risk being discovered by HM Revenue & Customs as a result of sophisticated new computer systems.
A new tool known as "Connect" compiles data to automatically flag people who may have underpaid taxes, using information from Twitter and Facebook, as well as more traditional sources, such as the Land Registry and Companies House.
"If you've paid a suspiciously low amount of tax, HMRC will know full well if you are lying if the evidence is splashed all over Facebook," said Richard Morley of BDO, an accountancy firm.
Any inability to explain how you appeared in a "selfie" at an exclusive restaurant, for example, will be taken into account when HMRC decides to impose penalties on the taxpayer.
Tax dodgers who lie to HMRC when during an investigation face paying penalties of 100pc of the tax due - double the original sum they were meant to pay.
Those who underpay by £25,000 or more are also "named and shamed" by the Revenue - with the names and addresses of tax evaders publicised on social media and in the press.
"Right now, HMRC is keeping tabs on Facebook to build up a 'timeline' of people's lifestyles," explained Mr Morley. "Teams of analysts are checking posts and pictures online."
Currently only publicly-available information forms part of HMRC's Connect database – but new data-gathering powers means this could soon change.
The taxman can now force websites including Facebook, Instagram and Twitter to hand over private personal details of its members if it is deemed necessary to a tax investigation.
It has already trawled the account details of people who use online marketplaces, such as eBay, to clamp down on the so-called "hidden" economy – people making a profit without declaring their income.
"HMRC can't simply pry out of voyeurism. There must be a case that the data is required to verify someone's tax position, and may need to be approved at a tax tribunal," Mr Morley said.
2) Pricing your car cover, health and home insurance
Insurance companies which snoop on customers' social media profiles are being investigated by the FCA's inquiry into the way firms use someone's internet activity to determine the price they pay.
The City watchdog said it might not be "all bad news" for customers because they could be placed into a lower-risk category due to their online presence, and pay cheaper premiums.
In a statement on the investigation into home, motor and travel insurance, the FCA said: "We are in the very early stages and at the moment the focus is on gathering as much information as possible."
Tracey McDermott, the FCA's director of supervision, said: "Social media is already an important tool for industry to engage with customers and its use is only likely to grow."
Telegraph Money has spoken to a firm that provides risk-scoring technology to major UK insurers, California-based Social Intelligence.
Unlike with HMRC, the firm won't look at individual tweets but uses "metadata", comparing you to people who show a similar level of social media activity, and how risky they are.
The firm combs through several "petabytes" of data. To give an idea, one petabyte is equal to one "quadrillion" bytes of data, or roughly 7,000,000,000,000 tweets.
Max Drucker of Social Intelligence said: "Our data shows us that a person with more than 200 LinkedIn connections, an email address that’s been active for more than five years and a private Facebook profile is statistically the least risky type of customer."
Statistically these types of customers will be less likely to make a claim, he said.
3) Detecting fraudulent claims
Facebook posts, Twitter activity and your "online footprint" form an increasing part of detecting fraud claims that cost the insurance industry an estimated £2.5bn a year.
Rob Townsend is a claims director at Aviva, Britain's biggest insurer, and said the firm detected 14,000 bogus claims totalling £95 million last year – many of them thanks to online tracking.
Facebook recently helped expose one £100,000 false whiplash claim concocted by 14 men who were all colluding.
They said they suffered whiplash after a collision between a car and a minivan.
But investigators became suspicious when they found that the driver of the car and the men in the minivan were all friends on Facebook. They later used CCTV footage to prove the crash hadn't even occured.
"People think insurance fraud is a victimless crime, but in cases like this fraudsters are staging accidents," Mr Townsend said. "We use sophisticated data analystics to identify such spurious claims."
4) Getting a credit card, loan or mortgage
Private banks use Google to research into the background of mortgage applicants, although this isn't yet routine among the major High Street lenders, experts say.
Adrian Anderson, director of mortgage broker Anderson Harris, said: "As soon as a new client is introduced to us we will Google them to learn as much about them as possible.
"You can learn a lot about people on the internet and even more through social media. The problem is that social media profiles may not always paint a true picture of somebody’s lifestyle."
People with no credit history could potentially benefit if lenders use social media scoring, argued Mr Drucker of Social Intelligence.
"There are millions of 'unscoreable' consumers out there. This group includes immigrants, recent university graduates and even retirees who have little to no credit history," he said.
Lenders currently decide whether to approve someone's application for a mortgage, loan or credit card based on their previous borrowing history, summed up in a credit profile.
There are three agencies which rate your credit: Experian, Call Credit and Equifax.
But "unscoreable" people, who may be reliable borrowers, can face paying higher rates of interest – or not being approved credit at all – due to their lack of a credit profile.
• The best of Telegraph Money: get our weekly newsletter