Desktop & mobile performances

Homepage speed of top half million sites

Httparchive is a non profit organization tracking websites. It may remind you Internet Archive which collects and permanently stores the Web's digitized content however its aim is different. This organization is  not interested in collecting and storing the digitized  content but rather interested in recording how digitized content is constructed and served.

Httparchive is a permanent repository of web performance information such as size of pages, failed requests, and technologies utilized. This performance information enables us to see trends in how the Web is built and provides a common data set from which to conduct web performance research.

The list of URLs Httparchive tracks is based on the Alexa Top 1,000,000 Sites. They are tracking only half of this list, half million URLs because as explained on their discussion platform, they run their tests  twice a month and they can only deal with half million URLs data in that time interval. The list of URLs is fed to their private instance of WebPagetest on the 1st and 15th of each month.

From the FAQ of the site we learn that  their test agents are located in a data center in Redwood City, CA in United States. Each URL is loaded 3 times with an empty cache ("first view"). The data from the median run (based on load time) is collected via a HAR file. The HTTP Archive collects these HAR files, parses them, and populates their database with the relevant information. The data is also available in Big Query.

Recently Google announced using page speed in mobile search ranking therefore I was willing to conduct some data analysis on mobile pages performances. However after a tweet post, I understand the need to include the desktop performances as people are interested in comparing them, desktop and mobile performances.

I downloaded the data sets of both desktop and mobile performances tests dated 15/01/2018 from Httparchive.

I added following type of information to each URL in the data sets.

{ "subdomain": "www", "domain": "google", "suffix": "com", "code": "US", "org": "Google", "dma_code": 807, "area_code": 650, "metro_code": "San Francisco, CA", "postal_code": "94043", "country_code": "US", "country_code3": "USA", "country_name": "United States", "continent": "NA", "region_code": "CA", "city": "Mountain View", "latitude": 37.41919999999999, "longitude": -122.0574, "time_zone": "America/Los_Angeles"}

Below you can find the desktop and mobile performances descriptive statistics of number of sites hosted in each continent and country  (top 10 countries hosting number of sites)  from httparchive data on metrics TTFB, fullyLoaded and SpeedIndex in milliseconds.


There has been a discussion on twitter about the min TTFB on descriptive statistics table of mobile performances which is 904ms. Patrick Meenan (creator of webpagetest, chrome engineer at Google) kindly answered the question and explained that for mobile 904ms TTFB as minimum is normal. He liked my tweet as well! You can find below my tweet and the following responses.

Desktop and mobile performances

Geographical view


Desktop TTFB

Mobile TTFB

Desktop fullyLoaded

Mobile fullyLoaded

Desktop SpeedIndex

Mobile SpeedIndex


Desktop TTFB

Mobile TTFB

Desktop fullyLoaded

Mobile fullyLoaded

Desktop SpeedIndex

Mobile SpeedIndex

Desktop and mobile performances

Top brands

If you are curious about mobile performances of some top brands such as Amazon or Google, you can find them below.

The domains are chosen by their domain name without tld. All the google sites are hosted in United States whereas Amazon's sites are hosted in 5 unique countries,  11 out of 20 are hosted in United States.

Desktop Amazon

Mobile Amazon

Desktop Google

Mobile  Google

I hope this data analysis gives you some information about desktop & mobile performances in general and you get some ideas about what may be the min, max, mean or median of important metrics for your mobile performances tests.

Thanks for taking time to read this post. I offer consulting, architecture and hands-on development services in web/digital to clients in Europe & North America. If you'd like to discuss how my offerings can help your business please contact me via LinkedIn

Have comments, questions or feedback about this article? Please do share them with us here.

If you like this article

Follow Me on Twitter

Follow Searchdatalogy on Twitter


Legal Terms Privacy

Python SEO

Learn Programming for SEO on Dataizer: Learn Python for SEO

Data SEO

Walid Gabteni: Consultant SEO

Vincent Terrasi: Data Scientist SEO

Remi Bacha: Data Scientist SEO

Recent Posts

SEO data distribution analysis 4 years, 10 months ago
87 million domains pagerank 5 years, 7 months ago
SEO Forecasting 5 years, 9 months ago
SEO data analysis 5 years, 9 months ago
BrightonSEO conference 5 years, 9 months ago
HTTP2 on top sites 6 years ago
Desktop & mobile performances 6 years, 5 months ago
Alexa top 1 million sites 6 years, 5 months ago
1 million #SEO tweets 7 years, 5 months ago
SEO, six blind men & an elephant 7 years, 6 months ago
Technical SEO log analysis 7 years, 8 months ago
3 ways for free https 7 years, 9 months ago
Crawl dictionary 7 years, 9 months ago
Https on top sites 7 years, 10 months ago
SEO web server log files 7 years, 11 months ago
Hsts on 7 years, 11 months ago

Recent Tweets