# The growth of HTTPS requests

**URL:** <https://discuss.httparchive.org/t/the-growth-of-https-requests/421>\
**Category:** Analysis\
**Created:** [November 19, 2014, 5:24pm UTC](https://discuss.httparchive.org/t/the-growth-of-https-requests/421 "2014-11-19T17:24:33Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![deanhume](https://yyz1.discourse-cdn.com/flex035/user_avatar/discuss.httparchive.org/deanhume/32/396_2.png) [@deanhume](https://discuss.httparchive.org/u/deanhume)\
**Post date:** [November 19, 2014, 5:24pm UTC](https://discuss.httparchive.org/t/the-growth-of-https-requests/421/1 "2014-11-19T17:24:33Z")

</div>

I am interested in working out if the number of requests over HTTPS have increased / decreased over the past few years and what trends I can associate with them. I’m using the following basic query to start off with:

```
  SELECT count(pageid) as NumberHttps, 'Nov-2010' as CreatedDate 
  FROM [httparchive:runs.2010_11_15_requests]  
  WHERE (url LIKE ("https%")) 
  GROUP BY CreatedDate 

```

While this request works great, it doesn’t take into account redirects to HTTPS. For example, a site may have a URL of [http://acme.com](http://acme.com), but actually get redirected to [https://acme.com](https://acme.com).

I did have a look at the following question - [http://bigqueri.es/t/how-many-https-websites-are-using-mixed-content/244](http://bigqueri.es/t/how-many-https-websites-are-using-mixed-content/244), but wasnt sure if this is along the right lines!

Any advice would be appreciated!

---

<div class="post-metadata">

**Author:** ![stevesoudersorg](https://yyz1.discourse-cdn.com/flex035/user_avatar/discuss.httparchive.org/stevesoudersorg/32/1293_2.png) [@stevesoudersorg](https://discuss.httparchive.org/u/stevesoudersorg)\
**Post date:** [November 24, 2014, 7:38pm UTC](https://discuss.httparchive.org/t/the-growth-of-https-requests/421/2 "2014-11-24T19:38:22Z")

</div>

That seems fine. I might use “count(requestid)” since the pageids are NOT unique.

One issue is the HTTP Archive assumes all the URLs start with “http://”, whereas all-ssl websites probably open in https:// (from Google Search, etc.). So the absolute # of HTTPS requests might not be too telling.

A better stat to track IMO would be the number of http websites that redirect to https.

---

<div class="post-metadata">

**Author:** ![deanhume](https://yyz1.discourse-cdn.com/flex035/user_avatar/discuss.httparchive.org/deanhume/32/396_2.png) [@deanhume](https://discuss.httparchive.org/u/deanhume)\
**Post date:** [November 25, 2014, 9:59am UTC](https://discuss.httparchive.org/t/the-growth-of-https-requests/421/3 "2014-11-25T09:59:02Z")

</div>

Great - thanks for the response @stevesoudersorg!

Your suggestion does seem like a better approach. Out of interest - is that how the HTTP Archive calculates it on the trend page?

[http://httparchive.org/trends.php#perHttps](http://httparchive.org/trends.php#perHttps)

---

<div class="post-metadata">

**Author:** ![stevesoudersorg](https://yyz1.discourse-cdn.com/flex035/user_avatar/discuss.httparchive.org/stevesoudersorg/32/1293_2.png) [@stevesoudersorg](https://discuss.httparchive.org/u/stevesoudersorg)\
**Post date:** [November 25, 2014, 5:17pm UTC](https://discuss.httparchive.org/t/the-growth-of-https-requests/421/4 "2014-11-25T17:17:44Z")

</div>

The HTTP Archive just looks at https URLs over total URLs.