As we head into the holiday season, now is a perfect time to pause and do a quick performance check up on your most important pages. The last thing you want is your high-value pages suffering from below average performance when exposed to peak season traffic.


Google Analytics makes this very easy with the built-in Page Timings report.


You can generate the Page Timings report in 4 easy steps:

1. Go to Behavior

2. Select Site Speed

3. Select Page Timings

4. Export the data (be sure to export all rows)​


Depending on your traffic and frequency of changes to your website you should export 1-4 weeks of data.



The Google Analytics Page Timings report lets you perform a quick analysis of individual page performance. For each page on your site, you’ll see an average load time based on sample of page loads during the selected time frame. Additionally, this report provides the the Page Value for each page; Page Value is intended to give you an idea of which page in your site contributed more to your site's revenue.


Using your Excel export of the full data set you want to find your pages that are slower than average and that have a high page value. These are the pages, or category of pages, that you should assess and optimize.


1. Filter Avg. Page Load Time (sec) to only show values greater than the site average. You can also filter based on a business target or best practice such as 3 seconds.

2. Filter out pages with a low number of page views since these pages will likely have a low number of page speed samples. An easy cutoff is to only look at pages that are above the average number of views for your data set.

3. Sort your table descending by Page Value


Now you’ll have a list of all of your pages that have below average page speed, above average page views, with the highest value page at the top of the list.


It’s very likely that the top pages on your list will be part of your shopping cart since these pages are involved in every transaction, it is also likely that you have very limited control over your shopping cart page(s) performance so don’t worry about those for now, focus on the pages you can improve with straightforward changes.  


For example if you discover that you have a large number of product pages with below average performance you can do an analysis of these pages to see if you can either compress or transcode your product images to yield better performance without sacrificing the visual quality and shopper experience.


Google Page Speed Insights is an easy to use, free tool, that will give you a quick assessment of the performance of any webpage along with optimization suggestions. Plug your URLs from your high value pages into the Google Page Speed Insights tool to discover why they may be slow.


With Page Speed Insights, you may find that your category pages have extra JavaScript or CSS that can be minified to yield improved performance. There are many factors that contribute to page performance so you’ll have to do a bit of sleuthing using Google Page Speed Insights or other tools to determine where you can have the most impact.


Additionally, if you want to make your Google Page Timings data even more useful you can increase your site speed sample rate above the 1% default. By collecting more samples, you will have higher fidelity page speed data. Here’s Google’s documentation on how to increase the site speed sample rate; it’s an easy change to your Analytics.js and you can collect up to 10,000 samples per day. An easy way to determine the initial sample rate for your site is by dividing 10,000 by your average site traffic and multiplying by 1.2 as Google Analytics tends to undersample.


We also have a helpful article covering four tips for improving website speed that can help you uncover actionable ways to make your web pages faster.


Now is the perfect time to take a few minutes and do a quick performance check-up and make sure your site is ready for the peak season ahead!


Stay tuned for Part 2 of this post where we talk about the difference between Google Page Speed data, Real User Monitoring (RUM) data, and Synthetic data for analyzing you webpage performance.