Core Web Vitals with Martin Splitt, Dave Smart, Giacomo Zechinni & Ulrika Viberg

Share:

Oh, what a night! We had an amazing evening yesterday speaking with some of the industries most looked up to experts including Martin Splitt, Developer Advocate at Google, Dave Smart, Technical SEO Consultant at Tame the Bots, Giacomo Zechinna, Technical SEO Consultant and Ulrika Viberg, Technical SEO and Founder of SEOGirls.

Core Web Vitals with Martin Splitt, Dave Smart, Giacomo Zechinni & Ulrika Viberg

A huge thanks to our amazing guest speakers and our hosts; Sophie Brannon and Maria White who are both Senior SEO Managers at Absolute Digital Media. We’re already counting down the days until our next #AbsoluteTwitterSpace… but without further ado, here’s more on the topic of last night’s discussion: Google’s Core Web Vitals.

Watch or read the episode below

 

 

Maria:

“Hi everyone, my name is Maria White and I’m a senior SEO at Absolute Digital Media. It’s my absolute pleasure to have in our panel some of the most knowledgeable people in Core Web Vitals. So welcome to our second Twitter Spaces chat, this one today on Core Web Vitals.

So, let’s crack on then with our event. On the 28th of May 2020, Google announced the blog evaluation page signal to better the web. This page signal measures elements of how users perceive the experience of interacting with a webpage. Then, on the 27th of May 2020, Google announced an update to search Console. It was then that the speed report became the Core Web Vitals report.

Each of the Core Web Vitals report, which are (import delay, largest content full paid, and communitive layer sheet) reflects the real-world experience of a critical user-centric outcome. So, ahead of its rollout in June, our chat today is all about Core Web Vitals and some of the most asked questions around them. So, let’s start with our panel.

We start with Dave Smart. He is the Technical SEO at Tame The Bots and Google search centre platinum product expert. Then we have Ulrika Viberg, she is Technical SEO and founder of the SEO girls in Sweden. We also have Giacomo Zecchini, Technical SEO Director. And from Absolute Digital Media we have our lovely ‘Rising Star’ Sophie Brannon. Finally, we are absolutely delighted to have Martin Splitt, Developer Advocate at Google.”

01. Maria: “Why do I see different values for the Web Vitals in Chrome user experience and paid speed insight or in the Lighthouse Report?”

Ulrika:

“As I have understood, the difference is that Lighthouse is actually using lab data and Chrome report uses field data. The difference between those is lab data measures how the browser renders the page or is loading the page in regards of what device you’re using, location and Bandwidth etc and if you have a lot of extensions or interference with the loading. That’s going to give you a different kind of metric than the field data which is actually measuring your users experience and then combining the experiences over a month. And of course, this will give a steady metric so to say.

Dave:

“Particularly Lighthouse when you want to develop a tool if it’s rare on your machine or your connection although it tends to simulate what a user on a mobile, certainly an approximation because it’s kind of stable it has to pretend to be something, and it’s almost designed to be fairly repeatable – it’s always same kind of connections to an extent.”

Martin:

“But also remember that the goal of all of this is to quantify the experience that people are having with your website. For instance the largest, for paid, that’s not necessarily always going to be the same thing depending a little bit on how the site knows how different things pop into your website and that depends on the network, latency retries. So, a lot of these things have inherent jitter, like a slight variation in themselves, so even if you would run lighthouse from your computer five times, you’d likely get different values within the same ballpark within a reasonable region. The biggest differentiation between that is lab data versus field data.”

02. Maria: “Martin, I have a question. Is it normal to see the scores in a website where they fluctuate erratically (30 or 40 points) from one day to another?”

Martin:

“You’re not the first one I hear this from so this can obviously be influenced by lots of different things so for instance if there is a program on your computer on one time running some kind of update taken up some of your internet bandwidth then you might see some sort of slowdown or if there is something that takes a lot of CPU in the background then that might take a lot of slowdown but normally, they shouldn’t fluctuate this much. The browser’s main purpose however is not to mention these things but instead to load stuff and then at certain points in time give off these metrics, these signals, to the metrics connection that is Lighthouse that basically allow you to measure these things. And they can definitely differ based on lots of different things. Any website will sometimes load a little faster or a little slower depending on Network congestion, but they shouldn’t fluctuate so much. If you can recreate this or reproduce these issues, try and open a bug with Lighthouse and ask why you’re seeing fluctuations. Lighthouse is very good at what it does but it’s far from perfect.”

03. Maria: “Why are the scores between lab data and field data so different if the lab data simulates the real scenario?”

Giacomo:

“The difference between these two is the amount of scenario that you can have. Field data can have tons of scenario; so, lots of people entering a connection with you or people everywhere in the world with good cable connections or poor devices all the way across the world. So, you can’t recreate all the real-life scenarios in lab data. You can have some things that can be grouped but we can’t replicate everything that we have in the field data.”

Dave:

“With field data, it’s looking more than lab data. Field data is all kinds of different people on all kinds of different devices and their interacting; they’ve got a million different tabs open which is something that lab data can never really get or simulate for everyone so there will always be some sort of discrepancy.”

Martin:

“In general, you get very different data from field data as well because multitude of users but also the aggregations but also look different and it’s a sliding window over time so you might have a month with unfortunate users and then next month you may see more users on more modern devices. So, this data is loud.”

04. Maria: “What are the new plug-ins, designed to change Core Web Vitals, doing to my website?”

Sophie:

“Certain plug-ins are marketing themselves not as a solution to core web vitals but it’s going to get your speed up there and we’re going to use all this cashing and functionalities. Marketers are jumping all over these which is great and like you said Maria, we’ve conducted some internal tests and we have some improvements in the Core Web Vitals metrics but is it just like a band-aid situation where it’s only solving some of the short-term issues and perhaps a more manual approach is needed.”

Ulrika:

“A lot of web owners don’t have a full stack of web developers that they can ask for help. In that case I think it might help but I don’t think it’s a silver bullet, none of these plug-ins ever are. You have to do something. You need a good web hosting service, and I don’t think the plug-in will actually do that. What do you say Martin?”

Martin:

“I know that very few of the projects I worked on in the past had the same path to success in terms of performance. I’m really excited about these plug-ins on one hand, specifically for people who want to show off the art or their small business, but the thing I’m worried about is handing over the housekeys to a piece of third-party software. All these services now have full access to what is happening on your page – at least on the client’s side. If they are doing something problematic, you have just added a layer of legacy instead of solving the original problem. You need to be aware of an additional complexity that may not work forever.”

Maria:

“Very, very interesting. It’s giving me an idea that every quick fix in SEO throughout the years has already been done. So, it might not be best practice.”

05. Maria: “So, I have a question from Joe Johnson, one of our Senior SEO Managers at Absolute Digital Media.”

Joe:

“I’ll jump in! Hi everyone, Joe here from Absolute Digital Media. A question for Martin, which links into what you were speaking about just now in terms of the PSI score and Core Web Vitals themselves. Which is more important; the PSI score or passing the Core Web Vitals? I understand the two are very much linked together, but I was messing around with the calculator on PSI and wanted to hear your thoughts.”

Martin:

“So, just to be very clear on this, neither lighthouse nor PSI which in turn is using lighthouse is actually relevant to the data experience signal. What they’re actually doing is giving you a way to test your pages. It’s more important to pass the Core Web Vitals – Specifically, we are using the field data for the page experience signal, so you want to monitor this data that shows up.”

Joe:

“My question that leads on from that is would the score itself be used as a tie breaker if you were to have two sites that were very similar, and both passed the Core Web Vitals. Would that PSI score be a tie breaker there, or is it still just about the Core Web Vitals?

Martin:

“Again, Google Search does not use PSI at all. It’s our testing tools for you to test things and to try things out and to measure things like in staging and development in production as well, but they are not used by Google Search.”

06. Maria: “Thank you so much! Another question here for Giacomo. If I build an AMP page, would I need the recommendation threshold?”

Giacomo:

“AMP helps you to produce some sort of pages that are fast. Through adding more script and functionality, you’re naturally going to have a slower page. But, through Using HTML, JavaScript etc. you can build a better website using AMP. Sometimes, using AMP you can already meet the threshold, but sometimes not. It’s not something you can be certain of. It depends!”

Dave:

“Yeah I think it’s fair enough to say that AMP can do a lot to help you when it comes to making the right decisions to build a fast page. But, as Giacomo says, it’s not a magic bullet.”

07. Maria: “Thank you. So now this question is more for Martin I believe. In your last Search Off The Record, you mentioned fake speed. What does that mean?”

Martin:

“Let me reemphasise that in our last Search Off The Record we didn’t try to actually speak about what is happening in Google Search, page speed or whatever. We meant that, if we were to make a new search engine today, how would we do it? And what practice would we look into? If we were to look into speed and performance, how would we do it? I reflected a question into that conversion from which I received multiple times on Twitter – What if someone tries to hurt me by making sure that there was a lot of field data reported back that makes my page look bad? That’s what they mean when they say ‘fake speed’. What if someone tries to manipulate that data – is it a genuine concern? To which I believe the answer is no.”

08. Maria: “Another question, again on page experience. Cookie pop-up banners. Do they count as intrusive interstitials?”

Sophie:

“I guess that depends on how intrusive it really is. For example, if it’s a pop-up that blocks all of the content on the page and it’s not responsive then yes maybe it will count as intrusive. But if it’s something that you can genuinely get around and users can interact with, then no it shouldn’t be at all.”

Dave:

“Yeah I don’t think it’ll necessarily be a negative!”

Giacomo:

“Please don’t use only images when you have a pop-up that is asking for cookie user information or something similar. From what I understand, Google needs to know what is inside the banner. If you’re using only images, Google can’t translate it. Write down that you’re writing for permission.”

Martin:

“Plus one to all that has been said!”

09. Maria: “Excellent! I have another question on that as well. Are automatic chat pop-ups bad practice?

Dave:

“I think it’s the same as covering everything, it can be intrusive in itself.”

 Giacomo:

“There are some chats that are loading everything together when you have the plugin on the page, but some that do not load everything. But for example, the images, if you hover over them with your mouse, they start loading everything. So, I would say no, it’s not a big problem if you have to using it as a marketing tool or anything else, but please be careful about the amount of data these chats are loading.”

 Maria:

“Thank you, that’s very interesting. I have a question that’s based on a comment once I heard from Martin. It’s about a polished turd – basically when you have an amazing, beautiful website with perfect speed etc, but the content is useless and not very good across other SEO elements. But, what about the other way round? If you have a website with lots of content, but the website in itself is a nightmare and the scores are very low. What would be the impact as a whole? I may be going into the territory of rankings, but it would be interesting to hear your thoughts!”

Ulrika:

“I had a thought about this, and we don’t know really how it’s going to impact our rankings because we’re not seeing it yet and so we don’t actually know what is going to happen. But I think that, as I see it, it will again be in its cluster compared to the other content pages within that site or cluster. Maybe overall scores are better anyway. But we should all try to optimise for a better experience which includes loading etc. Content itself is not enough.”

 Sophie:

“I completely agree. I think a lot of people are going to start going into panic stations and focus on getting their Core Web Vitals. Whilst that’s all well and good, we need to remember that Core Web Vitals and user experience is part of a much bigger picture. So, it does need to be a balance of having great user experience, page experience, Core Web Vitals, content, links, technical SEO. It needs to be a balance across the board.”

Dave:

“It’s about page experience. At the end of the day, the ultimate page experience is the content that’s on it.”

10. Maria: “Amazing, thank you. What would be your piece of advice looking ahead to the Core Web Vitals deadline which is coming in June?”

Ulrika:

“My advice would be to prepare and, if you haven’t started testing yet, start now and start optimising.”

Martin:

“On one hand we want everyone to be aware that web performance is important, but on the other hand, for many it won’t be very much of a big deal and won’t make much difference. It is tricky to balance this, and not be wrong or set wrong expectations, but there will be some people who will think that they should have worried about it sooner. On that note, try to do your work on how to reach your users and how to make them happy. If you do that, no matter how, you’ll be on a good path.”

Giacomo:

“If you can, look at cache.”

Dave:

Firstly don’t panic, and secondly have a search around for places that allow you to capture real user data. Get comfortable with the idea of the different metrics but don’t get too comfortable with them as they may change as time goes on.”

Maria:

“Thank you very much everyone, this has been interesting. The answers to all of the questions we had have been great. We received more than 70 questions in total! Big thank you to everyone who has joined us.”

11. Roll On The Next Twitter Space

There we have it! Some exclusive, insider knowledge about Google’s Core Web Vitals which are set to start rolling out as of next month. Another big thanks to all of our guest speakers and our hosts, Sophie and Maria, for yet another amazing session. More details about our next #AbsoluteTwitterSpace will be revealed very soon, so be sure to keep your eyes peeled on our Twitter (@AbsoluteDMedia) to be one of the first in the know!

For more information or to discuss your own digital marketing requirements, call one of our expert team today on 0800 088 6000.

Sophie Brannon
Sophie Brannon
Head of SEO

Sophie is our Head of SEO, who looks after the entire SEO team consisting of Account Managers and Content Account Managers. She provides support to the SEO team offering her read more.

Our Digital Marketing News

July’s Monthly Recap

What another crazy, action-packed month we’ve had at Absolute! We’ve been working hard to deliver you the greatest industry tips and tricks, took working from home to new heights, literally
Women working pool side on a laptop

Our Team Can Work From Anywhere, Not Just From Home

If there’s one thing we can all take away from the whole pandemic situation, it’s the power of flexibility. Around 4 months ago, Spotify announced its “working from anywhere” policy
5 actionable seo content tips for begginers

5 Actionable SEO Content Tips For Beginners

Businesses of all sizes have been there – searching on Google to find out where their products or services rank, and possibly feeling a little disheartened about what they find.

Start Now!

Join The Leading Digital Marketing Agency

Fill out the form to take the first step to digital success! Our digital marketing agency consultants are always on hand to hear from new businesses and answer any questions you may have. Select the services you're interested in and add any additional information about your digital marketing goals below, and we'll get back to you as soon as possible.


    Subscribe me to the latest digital marketing updates and promotional information