As full loading time does not actively capture the entire loading experience of the site (we'll explain more on this below), some of the key metrics we look at include:
- TTFB: Time to first byte: This referes to the time between the browser requesting a page and when it receives the first byte of information from the server. This time is inclusive of the time it takes to establish a connection with a TCP & SSL handshake (when request is made via https) and the initial DNS lookup.
- First Contentful Paint (FCP): when the browser first renders the initial bit of content from the DOM, indicative of when the user first receives feedback that "something is loading".
- Time to Interact: The time it takes before visitors can begin to scroll on a page, click on links, and interact with the website in general.
- Number of requests: It's important to compare the total number of total requests delivered from the 'original' website in China vs. the 'post-Chinafy' website in China. In most cases, this explains why the 'total page time' loaded may be greater than the original as the post-Chinafy site is delivering anywhere from 50-90% more resources.
- DOMContentLoaded: The DOMContentLoaded event fires when only when the initial HTML document has been completely loaded and parsed. This is often used to define when the page has completed loading fully. You'll notice that certain files will be deemed 'trailing resources', meaning that they tend to load near the end of the page loading experience. Examples of these trailing resources may include analytics (e.g. google analytics) which don't directly impact the user's experience of the site but are still nonetheless part of the site's content. In many cases, when the rest of the site is slow to load or has incompatible resources, these trailing resources and their functionality is negatively impacted because the site is unable to render them in a reasonable time.
- Page Complete: Chinafy defines this as the cutoff of page load time after the page load event is fired and there are fewer than than two resources loading.
2. Understanding Chinafy's testing methodology
On Chinafy side we run a combinations of tests on the site from various locations in China - some that are server-based, as this controls for user-variability (e.g. network, internet plans, human error etc.) and complement it with onshore testing as well.
- Real-Time Page Comparison: Comparing how your website loads across various locations in China vs. Chinafy preview
- Time Series Analysis: Comparing how your website loads in China vs. preview site loads in China over an elongated period of time. This provides clarity into how 'consistently' a site loads and the metrics pertaining to the average loading experience.
- Visual Speed Test: Comparing the visual loading experience across various locations in China
- 3rd-Party Testing: Webpagetest.org
- Manual Testing (aka "In-person" testing):Testing a site's experience from China.
*Note that we run these tests across a number of cities spread out across China including Beijing, Shanghai, Cheng du, Guangzhou and more.
3. How to go through the process of User Acceptance Testing (UAT):
From our experience, generally our clients (in this case you!), will validate functionality by reviewing the provided Chinafy reports, as well as what we'll also share from third-party tools like webpagetest.org.
- Is the Chinafy preview site loading more resources or less? You can review this directly in detail in the Resource waterfall.
- Is the Chinafy preview site interactive as intended?
- How is my site functioning? Are there parts of my site that were completely non-functional in China, and now functional? Pay particular attention to videos and images.
- How does my original site compare visually with Chinafy preview site?
- How does my original site compare in terms of consistency with Chinafy preview site? (A site that works 'sometimes', still doesn't work!)
5. Give us feedback!
Once you've onboarded onto Chinafy we continue to provide support on your site. Sites continually evolve, and similarly, we expect their optimisation process to be ongoing.
During this paid trial process, As you and your team will be most familiar with designed functionality, any specific feedback you/your team can provide on certain functions on the site would be most helpful (e.g. if there's a video that we didn't capture, or a missing image).
It's unlikely but possible, so this is a good occasion to understand what we can refine provided time to fully configure the site upon engagement.
^Support levels may vary between Chinafy plans. Please see more on pricing or ask your Chinafy contact.
Please sign in to leave a comment.