Feefo Compliance
At Feefo we have teams of people constantly on the lookout for anything that looks suspicious, irrelevant or inappropriate. These teams protect the thousands of businesses using our platform, the millions of consumers using our reviews, and our brand from a whole host of potential issues ranging from fake or incentivised reviews to illegal or fraudulent behaviour.
We use a variety of sophisticated tools and algorithms to detect and investigate potential compliance issues, some are fully automated and others involve in-depth analysis by our team.
Issues flagged by our automated algorithms are always verified by our specialists manually checking the business’s complete set of reviews to ensure current and historical data meets our quality standards.
Consumers can also raise concerns about reviews or businesses by directly contacting the team. These concerns trigger an investigation where our suite of automated and manual checks are completed.
If any compliance issue is detected, all reviews affected are immediately removed. This can include but isn’t limited to litigious, illegal and fraudulent content or behaviour. For most content related infringements, our system will moderate the review and offer the reviewer the opportunity to resubmit the review so that it meets our moderation guidelines; for fraudulent content or behaviour, cases are more complex so they will require a full investigation, after which our compliance team will contact the business to discuss their findings. In the event of any doubt, Feefo has a zero-tolerance for fake reviews. Our brand is built on trust and these rules are in place to protect our clients, consumers and our brand.
Moderation and Authenticity
Feefo Compliance involves two specialist teams:
- Moderation: This team focuses on the content of the review, for example, the words used or the inclusion of personal information. Full details as to why a review may be moderated can be found in our moderation guidelines.
- Authenticity: This team of fraud analysts dig deep into the data behind each review, looking at its source and its trustworthiness to ensure that there’s nothing that might suggest this is not from a real customer. As these processes are proprietary and helpful to fraudsters, we cannot share how this is done but we do publish some authenticity guidelines.
Avoiding Compliance Issues
As an invite-only review platform, our systems naturally have strong defences against many of the issues experienced by other review platforms. Only customer data from the businesses themselves is used to generate review requests, thereby reinforcing the integrity of genuine customer feedback. Only genuine customers of the service should be sent review requests.
Every business that joins Feefo agrees to our client terms and conditions confirming that they will send us genuine, accurate and complete customer data that hasn’t been manipulated or edited by them. This is where the authenticity team gets involved as they rigorously check sources and content for anything that might suggest a business may not comply with its contractual obligations.
Likewise, when submitting a review, consumers are also agreeing to our review submission terms and conditions. Reviewer non-compliance is primarily around the words used so this is where the moderation team take action, ensuring that such examples as unlawful, hateful and indecent content don’t get published. Authenticity can be a problem with reviewers too as the words used may be non-genuine, for example, copied from another website or created using an artificial intelligence (AI) engine. This is where the authenticity team apply their checks, detecting and weeding out anything that could be copyrighted or machine-generated material.
The checks for identifying authentic feedback are proprietary, complex and far-reaching so cannot be published however, compliant reviews are considered to be those with the following features:
- Have not been selected from the customer data for a reason that may prove beneficial to the business. Feefo clients are contractually obliged to send us an unbiased set of data. Therefore, there must be no pre-selection of which customer data they send us.
- Incentivised reviews with the correct label. Incentivised reviews are those where the reviewer was offered a benefit such as, but not limited to, money, commissions, discounts, vouchers, free products, free stays, invitations to events, favourable terms for loans/leases etc. to leave a review. Reviews which result in a charitable donation being made such as our product: Treefo or an entry into a prize draw (where the reviewer is offered only a chance of winning) do not have to be labelled. Contact support@feefo.com for labelling information.
- Specially labelled reviews from the business’s staff, friends, family or anyone closely associated with the business (only acceptable when labelled correctly). Contact support@feefo.com for labelling information.
- A customer should only receive one service review request per transaction, except where a genuine second service transaction has taken place. For example, initially collecting a review about the purchasing experience for a holiday immediately after the sale and then, after the trip, a review of the holiday experience.
- Have not been written on behalf of another individual. This includes, but is not limited to situations where staff leave reviews on the customer’s behalf, for example, if they don’t have internet access.
- Have not been collected at the point of sale, on the business’s premises or using the business’s electronic equipment (desktop computers, tablet PCs or mobile phones), with the exception of products such as Feefo Places and collection via QR codes printed on receipts. Reviews left onsite can be potentially biased.
- Have not been collected so they can be shared across multiple brands. Do not share reviews between brands. Reviews are collected for an individual website and therefore the same collection of reviews cannot be published on multiple websites, even those within the same parent business, as it can be misleading to consumers. Reviews collected via the Feefo syndication network do not apply in this instance.
- Have not have been created using AI, bots or any form of machine-learning technology. All reviews must be a genuine and first-hand account of an experience or product.
Action We Take
The compliance team is a data-led team in terms of its approach and depending upon the level of evidence, a number of actions may be considered.
Reviews that include issues listed in the moderation guidelines such as profanities, discrimination or competitor comments are removed as soon as they’re identified. Reviewers are given an opportunity to resubmit their review without the offending content and if the re-submission is considered compliant, the review is republished.
Authenticity issues often involve more than a single review and are usually considered to be fake reviews. Action will be taken immediately without prior warning to the business concerned with all non-compliant reviews being removed as soon as they’re detected. As this is a more serious example of non-compliance, in severe cases, further action may be necessary that include, but is not limited to:
- Turning off the feed that passes the business’s reviews to Google.
- Holding back or withdrawing any Feefo award such as the Trusted Service Award.
- Suspending the business’s use of certain Feefo products and features.
- Suspending the business’s entire Feefo account.
We will always contact the business as soon as we’re able to after the action has been taken. We submit our concerns, advise them of current and potential actions and give the business an opportunity to provide answers to our concerns. There are often perfectly reasonable reasons why we’re seeing problems so it’s important that we understand these in context. If the removed reviews are subsequently found to be compliant, they’ll be republished.