If you missed part 1 of this interview, please click here.
In this second interview, Bruce Zaayman, Performance Testing Director at Inspired Testing continues the conversation with Simon Hearne, Web Performance Architect, speaker and Blogger about Web Performance Testing.
Listen to the interview to find out more about the value of having a web performance strategy.
“3 seconds is a good goal to have for delivering all of your content to a customer before they get a little bit irritated and leave.”
- Simon Hearne, Web Performance Architect, speaker and Blogger.
Watch it here:
- Give us a little bit of an overview on yourself.
- What are the 2 or 3 questions that a client can ask themselves when deciding to embark on this web performance strategy?
- What is the general improvement that someone can expect when embarking on this web performance analysis journey?
- How do you build a business case to prove to your boss that web performance is something to consider?
- Expand a little but more about the balance of tooling and experience.
- Should people be worried about the state of performance on their sites?
- In terms of industry focus, is there a specific industry that should really focus on web performance more than the others or is it like retail, is it banking is it insurance or is it any and all of the above?
- We spoke about the fickleness of some of the customers out there. Can you give us some idea of the time it takes someone to decide to move on to another site or an application? Everything gets compared to Tik-Tok or Instagram that’s instant, so can you expand on that a little bit?
- LinkedIn = https://www.linkedin.com/in/simonhearne/
- GitHub = https://github.com/simonhearne
- Twitter = https://twitter.com/SimonHearne
- Website = https://simonhearne.com/
Full video script
Hi everyone, this is Bruce Zaayman, I’m joined today by Simon Hearne. This is Part 2 of a two-part series where Simon is exploring web performance testing with us.
Welcome Simon, I’m very excited to have this chat with you, let’s jump straight into it…
Give us a little bit of an overview on yourself.
I’m Simon Hearne, a web performance architect, and I’ve been working in front end performance for 10 years. I find real passion in helping clients improve the delivery of user experience, to get better business outcomes.
What does the future of web performance testing look like?
That’s a good question.
So, to think about the future we should think about the past, and over the last 15, 20 years, web performance kind of started as an industry and since then we’ve been working on better ways to measure and improve the performance of the sites and applications that we deliver.
We are constantly evolving the metrics that we collect and the ways that we collect them, so they have more value.
What I think is going to happen in the future is performance will move out of this technical space, the technical engineering bubble where it currently lives and move up into executive leadership and business strategy. The CEO and a CTO of any company should be able to quickly understand the performance of the experiences they are delivering through their website.
At the moment, I think we’ve been focused too much on technical metrics and bits and bytes and milliseconds, but with the future of better metrics such as core web vitals, we have a better communication method to share the performance of our sites and apps with more senior, non-technical folks and that I think it going to be a big shift in the way we approach it in the near future.
What are the 2 or 3 questions that a client can ask themselves when deciding to embark on this web performance strategy?
Yeah, I think the first one is “Do I know how fast I am?” and this is a good question because it’s almost impossible to answer, but by asking the question you start to think about where would I get that data from? What would the data look like? And how do I know what “good” is? Is it comparing to my peers? Is it comparing to industry benchmarks, or is it using Google’s goals for the SEO ranking?
And by asking that question I think you can get a lot of the way towards whether you need to work on performance optimization or not and once you know how fast you are you want to know how fast you should be and this is a tricky one. So traditionally we may look at competitors or peers in the marketplace and if you’re a retailer: are you fast as your peers and your competitors?
If not, then it’s possible that fickle customers, especially young customers that are mobile, may leave your platform to go to a competitor’s if you’re delivering similar products, because people don’t’ have patience to hang around and wait 4 to 5 seconds for a product page to load.
So once you know how fast you want to be, the next step is, how do I get there? And sometimes there are really easy steps to take like implementing a contact delivery network, like removing really big images from your product pages, but once you get beyond those quick wins, that low-hanging fruit, it becomes an architecture question. How do we help our engineering team to ensure that they are consistently delivering great, well formatted pages and apps to deliver the best possible user experience? And if you don’t have the confidence in that then don’t fear, a lot of companies don’t, and that’s where the support of someone like a web performance consultant or some assistance from Inspired Testing can help add that layer of performance understanding and expertise around your engineering team to help ensure the consistency of delivery of great UX.
That kind of leads me straight into the next question and it touches on a question we spoke about in Part 1, where we discussed proving the value of performance.
I know this is a loaded question and there are many caveats, but what is the general improvement that someone can expect when embarking on this web performance analysis journey?
Yes, and you’re right, it is a loaded question, and there are so many variables to consider: What are the metrics we’re trying to improve and what is the current state of play? But it’s not unheard of to get a 30 to 40 percent performance benefit by taking some of these steps towards improving the delivery of content. So even big companies like Yelp! Recently took on a performance project and improved their first paint time by 45 percent. And Yelp! has a big engineering team. So to start to look at your performance from a kind of outside-in perspective, rather than inside-out as you probably do at the moment; what are our customers experiencing? You can almost always find opportunities to improve, and we’ve seen sites improve by 70, 80 percent in the past, and just because things get worse over time if you’re not keeping track of them, so coming in, doing a quick, big bang analysis, a couple of sprints of engineering work to get back to a better baseline and then continual improvement after that.
How do you build a business case to prove to your boss that web performance is something to consider?
There’s no obvious, direct relationship between performance and business outcomes and revenue unlike implementing a new feature where you’ll have as part of that feature story “We will achieve X, Y and Z” in terms of revenue and user outcomes. With performance it’s more incremental change and if you improve a performance metric such as first content fill paint by 10 percent you can expect to see gains over time as users start to get used to having a better experience on your site.
Now there are ways to improve this over time. One of them is to make a change and measure the impact. So if you have an analytics platform, or a BI platform to collect the data about the speed of the experience it is your customers are receiving and their likelihood to do the things you want them to do to convert, to spend money, to have a large order value.
You can combine the two and say when we make things faster our revenue increases. Like I said, It’s not generally a linear outcome, a linear result but you can start to build that correlation over time. There are also tools such as real user measurement where you can track the experience that users are having in real time and start to segment that data. So take all of our users that have a poor experience and what is their likelihood to convert and compare that to all the users that have a great experience and determine what their likelihood to convert is. Often times you will see a really strong relationship based on the real users that are hitting your site, it’s kind of like testing in live and if you can get that data to show that a 1 second improvement in performance generally, correlates with a 5 percent improvement in conversion rate, that’s your ammunition, that’s the gold that lets you know how to improve and what the outcomes might be in the future If you achieve that performance improvement.
Thanks for that.
The one question that I do want to ask and I know that’s on the minds of a lot of people is the balance of tooling and experience. Can you expand on that a little bit?
Something we probably have a glut of in the web performance industry is tooling. If you ask someone to measure the performance of a website there could be a list of 20 or 30 or even 100 tools that can be used to measure performance. I guess the real thing is that understanding performance is important, but there’s a huge number of tools and different metrics and times that we can get from them, can be overwhelming and may be not even helpful in working out whether you’re delivering a good experience or a bad experience and then of course, how to improve it.
So I think the value in knowledge and experience is how to select the right tooling, select the right metrics and then use them to determine which goals we should be trying to achieve, because not all metrics are the same, not all tools are the same and what we need to do is find a balance of the right tools that give us the right data and not be overwhelmed.
So often we’ll speak to a customer that has 10 different teams measuring performance in 10 different ways and no one knows what the source of truth is. “How fast are we?” is a question that becomes really hard to answer. So helping to kind of narrow down and fix on a set of known good tools and metrics is kind of stage 1 in knowing how fast you are and how fast you want to be. And I guess it’s all the knowledge and experience over time and links with the industry to know which are the right tools to choose that help us deliver the value to customers.
In closing, should people be worried about the state of performance on their sites?
It’s something that can be a real risk to business success if it’s not understood properly and not optimized to deliver the best possible user experience. So it should certainly be up there in terms of the top priorities for the business especially moving into 2021 with the SEO updates that are coming and the difficulty of delivering a great user experience. When our customers expect everything to be lightning fast whatever device they use. This is a customer driven environment and we should be worried about how well we’re meeting our customers’ expectations.