This is just a un-edited parked page anyone having questions about the algorithm of our consumer segment. We are working on rebranding the site for our B2B. We have our ML models with associate a weight to 50+ characteristics of each page.

Fighting online scams is a 24/7 job, and we couldn’t hope to do it without the help of our sophisticated algorithm.

While we can’t reveal everything about the analysis taking place behind the scenes (otherwise the scammers would know too!), we hope to give you enough of an idea of how the Score is generated for any given website.

The Positive Indicators

Popularity

A website's Alexa Rank (no relation to Amazon Alexa) is a reliable indicator to judge the website's popularity based on the volume of website visitors in a given time frame. The popularity of a website can be a key indicator of how reliable it is.

Social Media Activity

Scam websites are often not too bothered by this, as it takes effort and resources to maintain social media pages. Therefore our algorithm can check if a site has active social media accounts and will adjust the Score accordingly.

Positive Reviews

We take into account not just reviews on our platform but also Amazon, Trust pilot and more.

Performance of the Website

Scammers time and effort is nt on directing users to their sites through spam, fake advertisements and malware. Therefore they can have sloppy, inefficient websites. Our algorithm takes this into account and will reduce the Score based on poor performance in this area.

Security of the Website

We also look at the security of the website; when the security cert was issued and when it expires along with any known malware on it.

The Negative Indicators

Based in a High-risk Country/Geography within a country

There are patterns related to the geographical location of that particular website. This does not mean that we blindly reduce the Score of sites from any particular region. We also take into account the factors mentioned here for the algorithm to evaluate.

Website Ownership

Genuine businesses usually have nothing to hide and do not hide their company data from the registration. Therefore, we take this indicator into account along with other factors to evaluate if the Score should be lowered.

Age of the Website

Scam sites are usually run as 'hit-and-run' operations, meaning that the site is created with the intention of being quickly shut down by the owners after scamming a lot of customers. Rarely do we see a scam site operating for more than three to six months. Scam sites don’t have a long shelf life as the sites get shut down within weeks by the owners themselves or due to other reasons such as users complaining, the ISPs (Internet Service Providers) or search engines blacklisting them, etc. Scammers know this very well and will set up new scam sites at a rapid pace. We take this into account and will check the age of an online store. If we can see that, alongside other scam indicators, a website has not been active for a long time, it will have a lower Score. As the website’s age increases, so will the Trust Score, assuming there are no other negative indicators.

High-risk Server

Every website has to live somewhere, and that home is called a server. But servers often don’t have just one website on it. Oftentimes, scammers will pick servers who do not mind questionable content. This means that these server hosts have a lot of content that is of the less reliable kind. Our Score can see the big picture, and track the scam sites that are clustered on one host. The algorithm can then reduce the score of future sites that decide to make those high-risk servers their home.

Ecommerce Platforms

Platforms such as Shopify and its Chinese counterpart Shoplazza give small businesses an easy way to set up shop online. On the flip side, the ease of creating websites provided by these platforms is exploited by scammers to create a large number of fake stores in a short time. Our algorithms take this fact into consideration and use it for determining the score

Pricing

We look at the mean prices of similar products across Amazon and others and if there is a significant deviation we take that into consideration along with other factors.

And More…

Of course our algorithm is complex and constantly evolving. These are just some of the factors that are at play when a website is assessed by our algorithm. We can’t give everything away here, otherwise, the scammers will catch on!