The Data

Our dataset consists of many Google Analytics accounts from contributors like you.

We anonymize all the profiles that our community has donated to the Benchmark in order to create an index of how e-commerce is doing right now. It works much like how the Dow Jones Industrial Average tracks the stock market.

How We’re Calculating Metrics

For most of the metrics that you’ll see in the Benchmark, we’re taking the average across all of the websites in our community and tracking the changes in those averages every 24 hours.

If you’ve joined the Tadpull Benchmark Community and added a site to the dataset, you’ll see the exact same metrics but calculated exclusively just for your site (and viewable only by you).

If you’re interested in more detail on the data analysis, you’ve come to the right place.

Percent Change in Revenue

For this, we’re taking the change in revenue for each day over the last week. Each day is calculated separately--the percent change for a day is based on what the revenue is for that day versus what it was the previous day.

For the benchmark value, we’re first averaging the revenues of all sites, then doing the percent change calculation.

So, sites with large and small top line revenue all get normalized to get an accurate sense of what’s happening.

Revenue Week Over Week

Average revenue per day for this week compared against last week as a percent change. Handy to see this at a bit higher level given days of the week can be a bit volatile.

For the benchmark value, we’re averaging this value across all sites.

Revenue Week Over Week

We’re taking the average order value per day for the last week, then averaging it for the whole week (Yes, an average of the average order value). We’re then comparing it against the same metric for the previous week and showing it as a percent change.

For the benchmark, we’re averaging it between all sites before we calculate percent change.

Site Traffic

We’re averaging users per day for this week, then comparing against the same metric for last week as a percent change.

For the benchmark, we’re taking the average of this value across all sites before we calculate percent change.

Conversion Rate By Channel

We’re taking the weekly average conversion rate for this week per channel, then comparing each channel against the same calculation for last week.

For the benchmark, we average the conversion rates across all sites before taking the percent change.

Users By Channel

We’re taking users per day for this week and totaling them up for the whole week per channel, then comparing each channel against the same calculation for last week.

For the benchmark, we average the total users across all sites before taking the percent change.

Bounce Rate

We’re taking the average bounce rate per day for this week and comparing it against the average bounce rate of last week.

The percent change calculated is percent change between the two bounce rates.

This may not be intuitive at first because at first glance, a bounce rate going from 30% to 33% is a 10% increase. And is actually worse than 30% (lower bounce rates are better).

For the benchmark, we’re averaging weekly bounce rates across all sites before we calculate percent change.

Pageviews Per Session

We’re taking average pageviews per session per day for this week and averaging them for the whole week, then comparing against the same metric for last week as a percent change.

For the benchmark, we’re averaging the average pageviews per session across all sites before we calculate percent change.

Average Session Duration

We’re taking average session duration per day for this week and averaging them for the whole week, then comparing against the same metric for last week as a percent change.

For the benchmark, we’re averaging the session duration across all sites before we calculate percent change.

Transactions By Location

For this map, we take all of the transactions in the U.S. and place them on the map, grouped by city. To reduce noise on the map, we also filter out locations that have too few transactions during this period.

For the benchmark, we total all the transactions across all websites.

Traffic Loyalty

For each day of the week, we take all users and new users. Then we total them up across the whole week and compare against the same thing for last week.

For the benchmark, we take the users for each day and total them across all sites.

Revenue By Device

We’re taking the total revenue for this week for each device category and comparing against the same metrics for last week.

For the benchmark, we total the revenues for each device across all sites.