Performance Benefits, SEO, & Advertising

[00:00:00]
>> Todd Gardner: Let's talk about the second reason why web performance is important, that's SEO Search Engine Optimization. How well you're able to help search engines understand and rank your content. If you're building a site that's publicly facing, you care about this a lot, you care about this because you live and die based on page rank.

[00:00:18]
Probably when the user goes to Google or whatever and searches for your website, your key terms, the things that you wanna rank for, you wanna be here. You want to be in that number one spot or as close to it as you can. Because if you're not in that number one spot, the traffic that you can expect to get drops off significantly.

[00:00:41]
According to Backlink.io, here is the traffic counts or click through rates by position and search results. And look at that number one, look at that number one. That number one result can get 75 million clicks on high volume words, and drops off significantly after that. The number two ranking is like 10% of number one, the number three ranking is half of number two, and it drops off really really fast anything time we get greater than 10.

[00:01:14]
So, if you can't get up to those top one, two, three positions, you're gonna get very, very low traffic. So great, you might be like, this isn't a course on SEO, and you're right, it's not a course on SEO. What the hell does this have to do with web performance?

[00:01:30]
Well, as of 2020, Google made some important changes that they announced on the Search Central blog. Search ranking this upcoming cert, or this change that happened in 2020, search ranking change that incorporates page experience metrics. They introduced a new signal that combines the core web vitals, which are an important measurement of performance that we're gonna talk about a lot.

[00:01:57]
They combine core Web Vitals with their other existing signals. So, as of 2020 the performance scores of your page are a ranking signal. You need to be fast to rank well, if you wanna get to that number one spot, if you're competing for a term like developer stickers or whatever, you need to be fast to get to that number one spot.

[00:02:21]
Or somebody else who has a good site, a site almost as good as yours, but faster will beat you out and the traffic wins that they'll get will be huge. Now, there's data to back this up, so here are LCP and CLS are two of the metrics in the core web vitals don't worry, you don't need to know what they mean yet, we're gonna talk about them a lot.

[00:02:40]
Along the bottom is the rank for common keywords, so the rank of the site and then what their scores were. And notice how much faster the number one spot is, the number one rank here is fast. Now, it goes up from there, there's this weird thing that happens around the end of the first page at like position six for LCP and position nine for CLS.

[00:03:07]
Where like the site gets faster and then it goes up again, and that has to do because this is not the only ranking signal. It also depends on like how good your content is, and how many people link to you and all the other things that go into SEO.

[00:03:19]
And so, if we assume that the top five pages are all probably equally good in terms of their backlinks and their content and the meaningfulness of how good of a site they are, but the first one is faster, the first one wins. So, you need to be fast if you wanna get to that number one position and there's huge benefits in terms of traffic if you can get there.

[00:03:43]
So, the third reason that's really, really important for web performance is if you do online advertising, and not everybody does, this might not be important to you. But online advertising is probably going to be super important for an e-commerce store or anybody that It sells something. So, when I say online advertising, maybe you think of this, you landed on some blog, some developer productivity site, some whatever.

[00:04:07]
And you went to a website because they have good content, and instead, there's ads smeared all over the place. Like here's some display ads that I made for developer stickers online and I'm covering so much of the page with ads, and you know what? Sometimes that's how the internet is, we're not here to debate the merits of online advertising.

[00:04:26]
But performance is a big aspect of this, and I'm talking from the point of view of the person who wants to do the advertising. Let's say I wanna buy an advertising campaign for developers stickers online. I make this really compelling ad that I want you to buy this, this sticker of a sloth on a rocket, snd so I spend $1,000 buying display ads.

[00:04:49]
Now for a consumer business, that'll buy me roughly 160,000 impressions, right? You don't really need to know a lot about online advertising for this but this is just some rough data. Now for a 160,000 impressions, that's gonna get me about 1% of them will click on that ad and try and buy that sticker.

[00:05:09]
So, if I have that website that I bought a $1,000, 1600 users, but my website tends to bounce 60% of the time, which means the user comes in and then because they're frustrated, they leave right away. They don't interact, they might not even wait for it to finish loading, they don't like the site, they're gone.

[00:05:29]
That means that at a 60% bounce rate, I get 640 real shoppers, I paid 56 to get somebody on my store. But 960 of them are just gone because of bad user experience, so how does this relate back to performance? Well, bounce rate and performance are directly related for a lot of websites.

[00:05:54]
In one particular case study by an online called Fernspace that sells furniture, they saw that with a 65% improvement in their web performance, they reduced their bounce rate by 20% and increased the time on page by 200%. So, by improving performance, they got users to not leave and stay longer.

[00:06:18]
Now, if we could apply the same sort of improvement to our developer stickers online site, I'm still gonna spend $1,000 I'm still gonna get 1600 users. That's just kind of general online effectiveness for this example, but I lower my bounce rate 20% to 48% bounce rate. That means I'm now gonna get 832 shoppers, that's a net of 192 new shoppers that would have come in that wouldn't have been there otherwise.

[00:06:47]
And I lowered my price from $1.56 per shopper to $1.20 per shopper. So, I can save a lot of money in online advertising spend if I'm able to improve my web performance. Now overall, we're talking about revenue, we're talking about dollars. Because if we wanna get something done, if we wanna invest in development work or web performance work, we need to make sure that there's revenue behind it to tie up.

[00:07:16]
So, there's lots and lots of examples on the web of how performance ties to revenue, here's an example from Walmart. Obviously, Walmart, huge site, huge fan base, huge traffic, but we can apply some of the lessons back to us. Now, Walmart did a big performance improvement, a big set of performance work.

[00:07:37]
And what they found is that 100 millisecond improvement in terms of load time resulted in a 1% incremental revenue for walmart.com. If they could speed up things by 100 milliseconds, so barely perceptible, but it caused enough people to not abandon what they were doing. To not have the attention span of a goldfish and not buy that Spatula that they wanted to buy online or whatever, they saw 1% incremental revenue.

[00:08:08]
Now, Walmart makes $665 billion dollars a year, 1% incremental revenue is 6 billion dollars, it makes $6 billion by that 1%. Now that's a scale that many of us are not operating at, but we can apply some lessons back to our sites. Now here's another example, Skill.co is an online job board, and what they found was that there was this direct relationship between load time and conversion rate.

[00:08:43]
A conversion for them would be like a job is successfully posted on their site, or somebody applies for a job. And at 2.4 seconds, they saw a 1.9% conversion rate, if that fell to 3.3 seconds, the conversion rate fell to 1.5%. If we get 5 seconds or lower, it's less than 0.6%, so the longer this goes, the worse the conversion rate got, and the less revenue that they would get.

[00:09:13]
There are so many examples of this, there's a site online called WPO stats, which is really cool to go and visit. You can kind of sort case studies by what metric they're improving or what business result they were after. And there's just so many examples of how different businesses have correlated improvements in performance back to their core business metrics.

[00:09:36]
And we can use this data to justify whether or not we need to spend money on web performance. Now, whether or not you need to spend money on web performance and whether or not it affects your business metrics is a complicated topic. And we're gonna talk about that a lot later in the workshop around setting appropriate goals for your site.