# Methodology

We strongly believe in being open and transparent about what we do. These are two key ingredients to build accountability and trust.

## Domains

Every day our servers process zone files for almost 350 million domain names. We have been doing that since 2013. With such an enormous amount of data we can construct historical domain trends pretty much for any web hosting provider out there with a high level of confidence. Our graphs updated on the biweekly basis. That’s the time necessary to process all zone files and update our database.

## Site Rank

Among web hosting providers, Site Rank value is inversely proportional to the volume of hosted domains. The more domain names a provider manages, the lower it’s Site Rank value. For example, GoDaddy with over 49 million domains has a Site Rank # 1.

Originally, Site Rank for review sites was based on the number of host ratings, customer reviews, Trust Score and data skewness. It was a complicated mixture and pretty hard to understand; hence, we have decided to simplify it. For now, to calculate Site Rank for reviewers we multiply the number of their ratings (hosts rated) by the Trust Score.

## Trust Score

Not all review sites are equal. Some of them have been built for the sole purpose of promoting specific hosting companies or generating affiliate income. They don’t bring much value to the community. That’s why we decided to give ratings on such kind of sites a lower weight.

Everyone starts at 5. The final Trust Score for each web hosting review site is calculated based on the following criteria.

Initial Trust Score | 5 |

Free to submit new hosts | +1 |

Ratings are based on customer reviews | +1 |

Reviews undergo verification process | +1 |

Vendors allowed to respond | +1 |

No affiliate links | +1 |

Data skewness | -|skew| |

## Global Rating

Since for each web hosting provider we have a set of ratings with different levels of relevance (Trust Score), we decided to use weighted average in our Global Rating calculations. We believe that is way more appropriate than using simple arithmetic mean formula, where you just sum up all of the ratings and divide by their number. Learn more about weighted average here.

## Calculations

Mean, Median, Mode, Standard Deviation, and Skewness are calculated using standard math functions in php. Nothing fancy, no addititional parameters. More info here.

- Mean. Calculated by dividing the sum of the values in a data set by their number.
- Median. The middle value in the data set. It separates the higher half of a data sample from the lower half.
- Mode. The value that occurs most often.
- Standard Deviation. 68.27%, 95.45% and 99.73% of the values lie within one, two and three standard deviations of the mean, respectively.
- Skewness. It is a measure of the asymmetry. Negative skew (left-skewed) indicates that the mass of the distribution is concentrated on the right of the figure. Conversely, positive skew (right-skewed) indicates that the mass of the distribution is concentrated on the left of the figure.