20Growth: How AI Will Change the Game For Content Creation and SEO, The Secret to Mastering SEO, When and How To Invest in SEO Most Effectively & The Best and Worst SEO Strategies with Joost De Valk, Founder @ Yoast

Joost De Valk is one of the OGs of SEO. As the Founder of Yoast, he scaled what was a side project plugin into a multi-million dollar business, used by 13 million sites and selling to Newfold Media in 2021. As one of the early SEO pioneers he is also an extremely coveted angel investor and invests through his company, Emilia Capital. In Today's Episode with Joost De Valk We Discuss: 1. From Side Project to Multi-Million Dollar Business: Why and how did Joost create the first version of Yoast? When did he realise that this was not a side project and could be a big business? What does he know now that he wishes he had known when he started Yoast? 2. When, How and Why To Invest in SEO: When is the right time to invest in SEO? How should one determine how much budget to allocate to SEO? Once decided on budget, what are the first steps to investing in SEO? Which part of the org should SEO team specialists sit in? What are the biggest mistakes founders make when investing in SEO/ 3. AI Changes The World of Context: How does AI change the way businesses create content? How can startups leverage AI to create and distribute more content for SEO? What are the biggest challenges/problems to leveraging AI for content creation? 4. Creating a Developer-Led Brand and Mastering PLG: What is the secret to creating the best developer-led brand? What are the biggest mistakes people make when marketing to developers? How does Joost navigate the balance between having enough value in a freemium product but also retaining enough value to be able to charge for the premium product? Is Joost concerned that budgets will revert back to CFOs and away from individual contributors with the financial downturn that is ensuing?

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  I don't think this is different from crypto. I think this has a bigger impact on society than crypto will. Because this has the type of impact that electricity and the steam engine had in terms of productivity changes.

• 0:12Harry Stebbings?

  This is 20 Growth with me, Harry Stebbings. And I'm so excited for the show today because the one element of growth we have really not done a deep dive on is the dark art of SEO. And so today, we're joined by one of the OGs of the SEO world, Yoast Devalk. As the founder of Yoast, he scaled what was a side project plugin into a multimillion dollar business used by 13 million websites and selling to Newfold Media in 2021. As one of the early SEO pioneers, he's also an extremely coveted angel investor, an investor of his company, Amelia Capital. And I do want to say a huge thank you to Shaq, Alex Schulz, and Matt Mullum Weg for some amazing question suggestions today without which this episode would not have been so special. But before we dive into the show today, I need your help. I need your input on the 20 VC Miro board

• 0:57Harry Stebbings?

  as I'm booking new guests that we should feature on the show in 2023. Now, this is the only way to make suggestions. Head on over to miro.com/20vc and leave your suggestions with digital sticky notes or comments. That's miro.com/20vc. And Miro actually sponsored this episode. For those of you that have not used it before, it's this incredible tool that I think is truly game changing. Just to answer 1 million plus people who sign up for Miro every month, it's a visual collaboration platform packed with the right tools, tech and templates to grow your idea, your product, your vision at scale. Miro acts as a single source of truth across teams. That means you can get buy-in much faster from teams like marketing,

• 1:40Harry Stebbings?

  engineering, sales, product, basically anyone who needs to have a say to make growth projects grow. And even though it seems like a simple digital whiteboard, it's so much more. Hashing out a sound acquisition strategy, brainstorm in Miro, rethinking your sign-up flow, using Miro's user research templates, quickly visualize team structures on a Miro board. However you need to grow, Miro can help. We use Miro all the time to brainstorm future shows, vote on potential guests, and leave notes for the rest of the team in our own time. Check it out again at miro.com/20vc to submit your guest suggestions. And I'm really excited about this next sponsor because a few of my portfolio companies use it and love it. So do you know the impact of your marketing efforts?

• 2:23Harry Stebbings?

  It's more important this year than ever to know the impact. Or how much you've spent so far this month and what you got as a result? You're not alone in not knowing these answers. Even though marketing is now all about the data, most companies still use manual spreadsheets for reporting and analysis. But if you want to supercharge growth, spreadsheets just won't cut it. Funnel is the world's first and only marketing data hub that takes all the manual effort out of data collection and lets you blend data from different platforms to unlock new, deeper insights. And the best part is this can all be achieved without writing a single line of code. I can hear your sigh of relief. That's why it's used by more than 2,000 of the world's top brands and media agencies, including Samsung, Adidas, Sony and Havas Media.

• 3:07Harry Stebbings?

  Start your own free trial by visiting funnel.io/20vc. That's funnel.io/20vc. And finally, we've all been there, but nothing is quite so depleting as the blinking cursor on an empty page. Reminds me of law school. You're sitting there trying to come up with ad copy or hypothetically a set of interview questions for a podcast you run, obviously personal, and it's just staring back at you like a brick wall. This moment of frustration amid growing content demands is probably why I'm hearing about Jasper non-stop these days. Jasper is an AI assistant for business that helps marketers, sales reps and other creators accelerate their writing process and repackage content into different formats, languages and tones.

• 3:50Harry Stebbings?

  Unlike broad use AI tools, Jasper was built for business and this makes a big difference. It incorporates SEO, a plagiarism checker and a browser extension. So you can call up Jasper in pretty much any text box or platform you use online. Visit jasper.ai/20vc for a demo. Your burnt out team will thank you.

• 4:16Speaker 00

  You have now arrived at your destination.

• 4:19Harry Stebbings?

  Yoast, I'm so excited for this. I told you, I was listening with Shaq last night, hearing stories. I've also heard from Brian Hale, from Malin Wedge, from Alex Schultz. So, thank you so much for joining me today.

• 4:29Speaker 00

  Absolute pleasure to be here. With all the people that have been here before me, it does feel like you're incorporated into this ridiculous group.

• 4:36Harry Stebbings?

  Do you know what? I'm very lucky to be ingratiated by such intelligent people. But I want to start at the beginning. So, you started the Yoast SEO plugin as a hobby. It's now used by 13 million sites, according to the trusty Google. What was the aha moment for like, I have to create the Yoast SEO plugin that even we use at 20VC?

• 4:54Speaker 00

  I started as an SEO consultant, having my own website, doing some small things. And I built a couple of small plugins that did various specific SEO things. And people kept saying to me, like, yeah, well, we just want one plugin that does it all. So, I just combined different plugins that I had into one thing. And then realized there's basically two sides to SEO, right? You have the technical stuff and you have the content side of things. And the technical stuff is not something that anyone wants to think about. What I tried to do is take all the technical stuff away from people and just fix it for them and give them good feedback on their content. Pretty simple recipe that has actually helped it grow so much.

• 5:31Harry Stebbings?

  But there's a moment when it changes from like a hobby that's kind of cool and kind of working to like, oh, wow, we actually have something. What was that catalyst for, I need to go all in on this? What was side hobby at the time?

• 5:42Speaker 00

  I started my own company in 2010, basically doing consulting. I was working for Alex, doing consulting for Facebook and doing consulting for eBay. What happened is that the plugin became really big, so it had like a million users. WordPress was growing quite fast and I suddenly had on this free plugin that I built as a hobby, a million users and I was spending way too much time supporting it. And my wife was quite clear, I don't need to make money doing this, or you need to stop doing it. I first started with an add-on, so I built a video SEO add-on, a small business plan for it. I was like, if I'm going to make a hundred K with this in the first year, then it's actually worth continuing doing this because I was making quite a lot of money as a

• 6:25Speaker 00

  As a consultant, so it had to actually offset that money. And then I made more than a hundred K in the first month. Oh, wait, there might be a bit more to this.

• 6:34Harry Stebbings?

  You have that moment though, then, as the next moment, you first have that catalyst for, oh shit, this is something that's working and working well. When you're doing a hundred K in a month pretty quickly, you're having people chase you down to invest. Why did you never take external funding?

• 6:47Speaker 00

  Why would we? First of all, don't think that any of this was a plan upfront. There was no plan to to sell the company. There was no plan to grow to as big as we as we became because it just happened. And we grew along with WordPress and we were very lucky in a way that we picked the right platform in the beginning and and built on that. But in the beginning, it was very much about turning that hobby into a bit of a lifestyle business. We have four children. I was also spending a lot of time with them. And I didn't necessarily want to build a huge business in the beginning. Then later on, when we realized that we had quite a large business in our hands by that time, it really wasn't needed anymore. We were generating enough cash to grow into what we wanted to do.

• 7:32Speaker 00

  The funny thing about our sort of business is that the only cost that you really have is people. Why take money?

• 7:38Harry Stebbings?

  If you could turn the clock back ten years or to two thousand and ten, when you started Ninkor Braces, what would you do differently?

• 7:44Speaker 00

  Had I known in the beginning that we'd be selling the company at some point, I might have hired a bit less people. A whole lot of the things that we did, we did just because we thought it was fun. And not necessarily because we were thinking about how to like make the most money. If I look back at it, then maybe we hired a bit too much at at certain points in time. And also, we're really not that big. Yoast's a company is not that big anymore. Thirteen million websites. Having a hundred and twenty or so people is really not all that much people.

• 8:13Harry Stebbings?

  I I want to start on the first major topic, though, because when we chatted before, you said, you know, alongside Matt, really, it's the first time we've had an open source and a kind of freemium OG on the show. And I want to start with something you said, which is open source slash freemium models are extremely hard to make work. But when they do, they lead to incredible potential, which VCs are underestimating. Let's break that down. Why are open source and freemium models? So hard to make work first.

• 8:38Speaker 00

  Two sides to that. On the one hand, freemium is always going to be hard because you have to decide about what to put in free and what to put in premium. Like, how do you attract the most users and at the same time get have enough in premium to actually convert them into paying customers? And the add on thing that's harder when you're in open source is that when people think that you've you've got a good product, but too much of it's premium is that they might just fork your premium product. And you also have to build like a sort of trust relationship with the customer. They need to trust you to be better than the other options that are out there. What you get quite quickly with these models is that you have web skill problems and small company income. At its peak, Yoast was a fifteen million euros a year revenue when we sold.

• 9:22Speaker 00

  That's not a whole lot of money compared to the eleven million sites that we were running on at that time. So you're solving problems for a whole lot of sites with relatively little people. At the same time, the open source community controls a large part of the web, a very, very large part. I think people still underestimate what Matt has done with WordPress. I mean, more than forty percent of websites on the web are WordPress sites. And companies like Shopify, and I heard you on a podcast with Toby, which was awesome. Shopify has like two, three percent of of the market, if that. So WordPress is way, way bigger and people underestimate that and it's under invested in, I think.

• 10:03Harry Stebbings?

  You mentioned there the challenge of putting enough greatness in the free that they convert to the premium. When you have founders come to you with this problem, I don't know what to do. I don't know how much to put in, how much secret sauce. What do you actually advise them? Because I'm sure you see so many.

• 10:17Speaker 00

  I will say that putting too much in free early on is dangerous because you can't really take it back. I think that's where open source slightly differs from from SaaS. Because in SaaS, you can just take it back. And in open source, as soon as you've put it out there, it's out there and someone is going to fork it if you take it out.

• 10:35Harry Stebbings?

  When you look at freemium and open source today that have worked, the best in class. What are the commonalities in those that have worked well that you think are core to their success?

• 10:45Speaker 00

  Underestimated thing in all of open source is branding and thinking about, hey, how do we want people to look at us? What are we putting out there? How are people perceiving us?

• 10:56Harry Stebbings?

  This is the hardest thing then, because you also, that's a developer-led brand. And that's what's the most challenging, which is like, I'm a branding guy, but a developer-led brand, that's hard. What makes a great developer-led brand?

• 11:09Speaker 00

  What makes a great developer-led brand is it's not just developer-led. If I'd been the only one building your SEO, then we would have only optimized for SEOs and for developers who want very specific niche features all the time to do very specific things. And honestly, nobody cares. I think what Brian, who mentioned in one of your podcasts at some point, that companies tend to optimize for the power user. I think that's the thing that you don't have to do, the exact opposite of what you should be doing. And it's very hard to do that well. So if you're just a developer leading a company like that, and it's really very important that you get someone on board who's not a power user, who can help you like, hey, I'm a normal user of this product.

• 11:48Harry Stebbings?

  When you think about don't optimize for power user, I'm fascinated there because quite often they're the ones most likely to churn. They're not the real lovers. I understand optimized for the power user and for the super lover, because they're your champion. They're the ones who are gonna espouse it to their friends in the community, make the ones who love it, love it more, and they'll do the rest for you. Why am I wrong?

• 12:07Speaker 00

  Well, you're not. That's where it's super hard. The fine line that I've always tried to do with yours is that we did the things automatically that the lovers, the really geeky SEOs really liked. So we fixed all the problems automatically and very cleanly. And we tried to keep the interface as simple as possible. And at the same time, we horribly failed at that at times. The thing is that once you start looking at some data, and one of the harder parts is that you actually don't have that much data, is that people don't look at settings. So for an SEO plugin, they turn it on and they expect it to do magic. That means that your defaults have to be incredibly good and that every setting you add is basically just more noise.

• 12:52Speaker 00

  So I think that the power is to make sure that you keep. On doing the right things and reacting quickly. So one of the things we always did was whenever Google came out with new stuff is we'd be on that. And we'd have a new release ready a couple of days later, or even on the day itself to add the features that were needed.

• 13:10Harry Stebbings?

  You mentioned kind of the default there and like settings, people don't go to settings. I think it was Dave Morin. Who said, the devil's in the default when it comes to products and product expectations. And then actually it just destroys creativity. Do you agree? Or is actually the default is good? It's what we know and it's what we're comfortable with.

• 13:27Speaker 00

  I think the default is what 90% of your users will end up using. So it better be damn good. It's actually one of WordPress's core principles in its design philosophies, which Matt came up with, decisions, not options. That means we're not going to add options for stuff where we can decide the right thing for 90% of people. We're not going to add an option for it. There might be a filter in the code to change it. So if you're a developer, you can change it. But for normal users, adding more options just makes your product harder to use.

• 13:57Harry Stebbings?

  Can I ask, I think we learn a lot from mistakes. When you look at kind of the product journey of Yoast. What do you think was the single biggest kind of product mistake? And what did you learn from it?

• 14:07Speaker 00

  The biggest product mistake is thinking that people were using our features when we didn't know for sure. So we spent a whole lot of time building features that then nobody used. And that's the problem of building a WordPress plugin, because you don't have telemetry going back to your systems all the time. So we literally didn't know. And then we started building tracking to actually send some telemetry back. And then we realized, oh, wait, we're spending a whole lot of our time working on this and nobody's using it.

• 14:40Harry Stebbings?

  So what do you change then? You spend more time on branding, you spend more time on distribution, you spend?

• 14:44Speaker 00

  No, you spend more time on the parts of the interface where people actually do interact with your software.

• 14:49Harry Stebbings?

  I do have to ask that. I'm concerned when we look at the freemium model in particular, especially as we enter a different macro climate. Because what you see is the removal of the credit card from the user and actually reverting back to the CFO where it traditionally was. And people don't have the same purchasing power on the front lines. Are you concerned in the same way as I am about that removal of purchasing power? And how does freemium look differently over the next few years to how it did over the last few years?

• 15:13Speaker 00

  I am slightly concerned there. I'm not as concerned for Yoast because SEO is probably going to be invested in more instead of less in a down market. Because people look at their AdWords bill and they go like, oh, maybe we should do something about that. But yeah, I think you're right. And at the same time, most of these freemium models are relatively cheap. So while there's not purchasing power as much anymore on the on the front end, I mean, Yoast SEO costs like 99 euros a year. A lot of WordPress plugins have those kinds of price points. Those are not the hardest decisions in the world usually.

• 15:46Harry Stebbings?

  What worries me also with the freemium models, honestly, and the kind of the cohort of startups we have today is you have the cost structure of enterprise startups. You have SDRs, you have AEs, you have CS teams, you have this very bloated kind of org structure. And then you have the price points of freemium and we have no knowledge that they upsell efficiently. We have no knowledge of like large ACVs at the end of the rainbow. And so you have this inflated cost base and this small ACV with freemium and it just doesn't doesn't work.

• 16:15Speaker 00

  Honestly, a lot of those companies are also building for scale that they're never going to reach.

• 16:20Harry Stebbings?

  Yeah.

• 16:20Speaker 00

  I think that there's a lot of opportunity to just do scrappy and small. And also to price reasonably because some of these things just get too expensive and then it's too easy to throw you away. So something like Calendly, which I see a lot of people use and love, Google rolls out a new feature and suddenly Calendly is gone.

• 16:40Harry Stebbings?

  Do you think so? I mean, it's a three to five billion dollar company.

• 16:43Speaker 00

  I mean, the risk of that is quite big, right? I mean, Calendly, I don't, I no longer pay for it. But when I used to pay for it, it cost me more than my Google Workspace account. And I was like, these pricings don't match.

• 16:53Harry Stebbings?

  Again, I'm an investor for a living. You said before that VCs underestimate these models. How do you look at them? How do you evaluate them? What is the mechanism of determination for quality?

• 17:05Speaker 00

  In terms of software, I can just look at the code. I'm arrogant enough to say that I think I can determine that myself. And then it's just looking at what users are saying, looking at reviews, looking at how people are talking about it. In these communities, what you really get is that you have like, the WordPress communities are, it's a layered. Thing, but the inside WordPress community is a relatively small group of people that goes to work camps and that goes to these conferences. And when I say small, that's relative, right? I mean, WordCamp Europe will probably be a couple thousand people. It's actually very cool to listen to that and figure out like, hey, these are the ones that are really going to make a change. So that's being in that community and actually caring about that community and being aligned with that community.

• 17:48Harry Stebbings?

  How do you tell pre-launch? I meet many, many in my role.

• 17:52Speaker 00

  Why would you tell pre-launch? You don't have to.

• 17:54Harry Stebbings?

  Well, why would you fund pre-launch?

• 17:55Speaker 00

  Yeah, but also why would you want to tell pre-launch? I mean, most of these things exist already. That's the, so it's a very different building atmosphere. It's very much built out in the open.

• 18:05Harry Stebbings?

  Not many startups actually take that approach, though.

• 18:07Speaker 00

  But that's what I'm saying. Like, this is very uncommon to, I think, most VCs, that it is much more out in the open. I mean, when you sell Yoast, it's what you sell is a brand and what you sell is not necessarily just the software, because the software everyone could just pick up and copy.

• 18:23Harry Stebbings?

  And do you think any VC does it well?

• 18:24Speaker 00

  So I've spoken to quite a lot of them because they do reach out and these VCs are going to VC. I don't think a whole lot of them get it. I think there's a wide open market.

• 18:33Harry Stebbings?

  Oh, what is chance for you to become a VC? As I said, everyone wants to be a VC these days. I want to talk about kind of the highs and lows of SEO. We met through Shaq, really. And he told me about the instrumental work that he did on CoinDesk. So, timing and budgets. What does it mean to invest in SEO? I often hear like, oh, we're going to invest in SEO. What does it mean to invest in SEO?

• 18:54Speaker 00

  I think that every founder should understand the basics of SEO because it's going to help them in a whole lot of different things. Because what you need to do when you do SEO is you need to look at a whole lot of data. And you need to think about what do people search for? How do I describe my product in words that people actually use? When they have a problem, what are they searching for? What are the words that people use to describe the problems that they have? And how can I tap into that? I actually think that that's good to think about when you're building a product. So investing in SEO starts there with actually thinking about what are we solving? Which problem of our users are we solving? That's where SEO is in the beginning and where it's important. Important, I think. And then when you invest in SEO, you need content and you need a good tech platform.

• 19:38Speaker 00

  And honestly, WordPress is out there. So if you're, if you're gonna use another CMS, I think you're pretty stupid. Because every other CMS is gonna cost you more money. And then you need content, you need to write that content. And in the beginning, that's probably you that needs to write that content.

• 19:54Harry Stebbings?

  If we go back to like the founder hat of like what it means to invest in SEO, the next question is like, when's the right time? A lot of founders like, Harry, I just need to get a couple of customers. I need to build the base and then I'll think about it. When's the right time to start investing in SEO?

• 20:07Speaker 00

  If you pick the right platform to build your website on and make sure that it's relatively technically SEO friendly already, then you're good in the beginning. And then as soon as you start doing customer support, you'll get questions that you have more than two or three times. Start writing knowledge based articles for those. It's as simple as that. The things you're already writing anyway, because you're answering those customers or you're fixing them, reuse that content and put it on your site, then just build on that. I think the right time to start investing in it is fairly early, but that doesn't mean hiring an expensive SEO consultant. It just means putting content on your site and making sure that people can find what you do.

• 20:46Harry Stebbings?

  You mentioned hiring an expensive SEO consultant. You see so many people attend SEO. What are the biggest mistakes you see founders make with SEO in the early days?

• 20:53Speaker 00

  One, come up with a brand that is not recognizable. Even in search, if people search through your problem area, they never search once, right? They always have like a search journey where they search for multiple words and for multiple things over a period of time. Even if they don't click your result, if they see your brand name a couple of times in their results, you're slowly building up confidence for your brand. So being recognizable in search is super important. People underestimate how important it actually is to have a brand that people can remember and that doesn't become very vague very quickly. That's one. The other is coming up with terms themselves for the, say, you think you're very clever and you create your own term for what you're doing. Nobody else uses that. So don't.

• 21:38Speaker 00

  Use the words that everybody else uses.

• 21:40Harry Stebbings?

  That's so interesting. So category creation may sound Sexy to a VC, but actually whatever intellectual word you're using, it doesn't mean shit to anyone. Don't bother.

• 21:50Speaker 00

  So category creation is, I think, super important, but it's almost always it's a category that already exists as a problem. The problem that you're fixing for people already has words attached to that. If you're going to start using different words for that category than the people are using to search it for, then you're gonna have a hard time getting people that search into your website. You can't really create demand with SEO. SEO is very demand driven, just like AdWords is very demand driven. The demand has to be there. You can't create the demand. So that's why the branding is important and you have to do other things as well. And a holistic view on growth should absolutely include SEO, but should also include a whole lot of branding.

• 22:31Harry Stebbings?

  Should you have SEO specialists within your growth team, within your marketing team? Is this a meta skill that everyone should have? How should that be distributed within your team?

• 22:40Speaker 00

  I think you should have SEO specialists on your team if you're building a larger growth team, because it's a pretty specific problem area to keep track of. It's quite a bit of time that it takes to just keep track of everything that Google does. As your company gets larger, managing SEO gets a lot harder. So you'll probably need an SEO team over time because.

• 23:00Harry Stebbings?

  What changes with time? Why does it get harder?

• 23:02Speaker 00

  As you move quickly on your site and all these companies want to ship every day, then it's hard to not make SEO mistakes every day. So you need people that keep track of what you're shipping and say, okay, maybe we can change this so that it's actually SEO friendly.

• 23:18Harry Stebbings?

  Who do the SEO specialists in your team work closest with? Is it with the product team? Is it with the dev team? Is it with, who are their consigliaries to ensure they have that seamless integration?

• 23:28Speaker 00

  Both these, actually. I think that they should probably be in that triangle the whole time. And the content side of things should probably also talk to your PR and comps, to your people, et cetera. SEO has a nasty habit of being in so many different places at the same time, like so many different things are important to the SEO.

• 23:45Harry Stebbings?

  You mentioned the centrality of content there. I think one of the things I always say with content, you know, having done it for years is the reason most people fail is because. They give up. It just takes a long time to work. When you think about how long it takes to see the gains of SEO, how long does it take to see SEO really come into fruition?

• 24:02Speaker 00

  This depends a bit on where you are in your journey, but if you're starting as a new company, then it's going to take quite a bit of time. Nobody's heard of you and you're not going to get like a whole lot of clicks automatically suddenly because you've launched a new website. At the same time, if you were a bigger company and you've not invested in SEO, but you do already have a brand, then you could see results very quickly as soon as you start actually putting decent, well optimized content on your site.

• 24:28Harry Stebbings?

  If you are a CEO or CFO of a series B company and you haven't optimized for it already, how do you allocate budget for SEO? Do I take it from marketing? Do I take it from product engineering? Is it a new budget? How much should it cost? If I was literally in an angel investment of your series B, what would you advise me in terms of asset allocation towards SEO?

• 24:49Speaker 00

  If you're already doing CPC, I'd take 10% of your CPC budget and allocate it to SEO. And then start from there and see what happens.

• 24:57Harry Stebbings?

  How do you measure along the way the effectiveness and success of your SEO strategy in those early days? How do you advise me on that?

• 25:05Speaker 00

  Well, not just early days, measuring the success of SEO is hard because Google has very successfully made it impossible for people to track the effect of anything but CPC. Every company out there either uses last click attribution or an attribution system where the last click is important. And the last click is in most cases going to be a brand search because you've convinced someone to buy somewhere in the process. We don't know which keywords people search for anymore when they get to a site because Google hides those. So it's actually quite hard to say, hey, how well does SEO work for us in our business? The same way that it's hard to tell the results of a TV ad.

• 25:45Speaker 00

  But that doesn't mean that TV advertising doesn't work, nor does it mean that SEO doesn't work. It really is something that's going to blow up the rest of your results and it might be hard to attribute. So actually that's one of the falsities that we have in the, in the whole growth world is that we can measure everything. Everything? I don't think you can.

• 26:05Harry Stebbings?

  When you go inside companies today, you're investing, you're advising, what are the best SEO individuals and teams have and do, and what are the worst have and do?

• 26:14Speaker 00

  I think the worst start looking at a very long list of technical things to fix and go from top to bottom and just spend months and months and months fixing things that really don't matter because the overall story is bad. The best go in and look at, okay, so what is our brand story? What is the thing that we're selling? They fix the biggest technical problems and then go into, okay, so how do we create a story around this? How do we, together with comms and others, make sure that people that land on our site do the right things? And find the content that actually addresses their needs.

• 26:47Harry Stebbings?

  So you said about brand story. We're going to do this in real time because there's going to be a load of founders that listen to this and go, great, I get it, I get it, but what does that actually mean for us? So Yoast, Yoast SEO for everyone, is that your brand story? Just so I understand.

• 27:03Speaker 00

  I think Yoast as a company, I can still do this better than we probably do. Yes, it is. I do think that SEO for everyone is the thing that Yoast does. It is making SEO accessible to everyone. But the things that people search for are terms like WordPress SEO. And if you search for WordPress SEO, you'll either end up on our plugin page or on an article page that gives you a very long list of things to do for SEO. And it explains to you how SEO works.

• 27:30Harry Stebbings?

  You know what you get? You get three ads.

• 27:32Speaker 00

  That's one of the funny things where you start selling your company, people start asking for things like your cost of acquisition per customer. And you were like, we never spent any ad, so how would you calculate this?

• 27:42Harry Stebbings?

  I think a final thing on this, which I do have to touch on, it's really hard though when you have a horizontal tool. Say you have a Notion, say you have an Airtable, say you have any of these kind of collaboration tools which are applicable across many different verticals. How do you think about an effective brand story that hits those very different use cases in one short headline?

• 28:01Speaker 00

  At that point, you're going to want to start optimizing the use cases. You'll have a ton of use cases. You want to tell that story. You want to tell the stories of how your customers are using it and writing that down. All the different ways in which they are using it and all the different verticals that you're thinking of where people could be using your product, you want to be creating pages for that and actually telling people those stories.

• 28:24Harry Stebbings?

  Should you do that from day one or should it be integrated over time?

• 28:27Speaker 00

  If you can do that from day one, it's absolutely awesome.

• 28:30Harry Stebbings?

  You mentioned the centrality of content again. We're seeing OpenAI change a lot about low-level content creation. How does AI impact search and SEO, do you think, most significantly in the next couple of years?

• 28:43Speaker 00

  I think AI for content generation is still problematic because it hallucinates too much. So there's too much things that are in there that are just not true. Predicts the next word, but it doesn't mean that the next word has to be true. At the same time, I think AI can be tremendously useful in content generation. It can, it can help you create outlines, it can help you create excerpts of longer articles very quickly. It can actually, can help you distill your research very quickly because you can say, hey, so I've got these thirty articles on this topic. And you can use something like OpenAI to make excerpts of those and summaries and, and use that to write your own content very quickly. But I do think that writing your own content and adding something unique is going to be more important over time.

• 29:28Speaker 00

  And with a lot of these changes happening, search engines are bound to focus more on the author of the content than they have done for a while. Because it's actually important for them to make sure that what's on that page is true and is good. They want to make sure that they can actually trust the author. I think the author of an article is going to be more important over the next few years than it has been in the past few years.

• 29:50Harry Stebbings?

  Does the price of content not go down? If you think about what this does to demand and supply chart, this is a supply shock, creating excess supply or a massive increase in supply, which will inevitably increase the price by which demand is willing to pay. Do you not worry about that?

• 30:05Speaker 00

  In terms of like more content being created, yes, there will be loads more content being created at probably very, very low prices. The thing is that most of that content will be not good. The content that will rank in search engines will very quickly be the content that is actually unique and, and actually has value in it that is not given by an AI solution.

• 30:28Harry Stebbings?

  Do you feel we lose creativity in our content teams? And what I mean by that is you mentioned like the outlines that it can generate. Is there a loss of creativity with not having to do this? It's like. You know, our arithmetic skills went down dramatically when calculators were introduced. Do we lose that?

• 30:43Speaker 00

  Some companies might lose that. I think the winners will not lose that. I think creativity is still what's going to make content truly good and truly unique and shareable. SEO doesn't exist in a vacuum. If you land on a page that was written purely for SEO purposes, it doesn't have any value. You're not going to share that page with your friends. At the same time, the thing that you want for pages that rank in search and for them to be successful is that people will share them with their friends and will revisit them. So we have to make better web pages. And Google has been saying for years, you need to make websites for users, not for search engines. And I think they're absolutely right. If that page that you land up on from search that you build for SEO is not something that I want to share with my friends, then that page is not worth its money.

• 31:28Harry Stebbings?

  You say about kind of, you know, not many people do it well, but the importance of quality of content. I agree, but what concerns me is when you increase supply, you have discoverability challenge now. Whereas it's just harder to find that great quality content because it's in a sea of rubbish content. That's what concerns me.

• 31:44Speaker 00

  That concerns me and I think that concerns search engines even more. That is one of the things that Google is panicking about. Like this is going to create a wave of very, very bad content on the web. And how are we going to surface the really good content? And that's why I expect them to be doubling down on things like the EA team, the things which they've had, expertise, authority, trust. They've already talked to that on quite a number of occasions and I think they'll double down on that.

• 32:10Harry Stebbings?

  What elements of AI over the next few years concern you that you don't think enough people are spending time on or thinking about?

• 32:17Speaker 00

  I think the biggest problem, and this is where it shows that I'm very European. Is that we don't know what's in those large language models and that we, we really should. And we sort of have a right to know as well, like what did you train your model on? What can I, based on that, I can make a decision on whether I want to use it for a specific use case. And I would be very much in favor of some regulation that actually say, hey, you, you need to disclose these things about your large language model in order to be able to, to allow the general public to use. Because then they can know, especially because a lot of those large language models are trained on content that they don't necessarily own. I I've had the the pleasure of working with the team at the Guardian for a couple of years about a decade ago.

• 33:01Harry Stebbings?

  So they should pay for access to that content, which then enables them to provide their service better. They're leveraging other people's work to make money off it for free.

• 33:09Speaker 00

  That is hard and I don't have the answer. And and part of that is going to be licensing, but part of that's actually also going to be S E O. Like, how do we create robots that T X T rules for this this sort of stuff that like allow you where you can say, I want to be in in this type of large language model, but not in that. That is going to be super hard. I don't think we have any good solutions to that yet.

• 33:31Harry Stebbings?

  How do content teams and the structure of them change with A I playing a much more central role in the creation and planning of their content that they do today?

• 33:41Speaker 00

  I hope it speeds them up. If they use it well, it speeds them up and you might need actually a few less people to do the same amount of work. There's very many ways in which A I can speed up not just content, but development too. If it doesn't mean a step changing your productivity, then you're using it wrong.

• 33:58Harry Stebbings?

  What worries me is actually like, everyone agrees that it's going to create many billion dollar companies with ten or twenty employees, largely engineers. What that does to wealth inequality is terrible, even worse than what we have today. Do you not worry that we just usher in a new phase of income inequality and wealth inequality that we've never seen before?

• 34:17Speaker 00

  Well, yes. So this is where I admit that I'm pretty much a socialist, which I know is a bad word to most Americans. I do worry about that. I do worry about the fact that that the wealth inequality we create is not something that we're going to be able to sustain. So we need better systems for that. And I also worry about like, how do we even make this understandable to politicians at some point? Are they going to be too late at the the point where they start regulating this, where we've already gone past everything that they could fix?

• 34:46Harry Stebbings?

  What is that point of being too late? When it's already in the hands of consumers and bluntly, there's riots on the streets when you take it out?

• 34:53Speaker 00

  That's a good question, and I don't have the answer.

• 34:55Harry Stebbings?

  I thought about it a lot with crypto, which is like, you can't just put. Crypto back in a bottle. If you could offer it to politicians at the start to put it back in, I'm sure they would take that.

• 35:03Speaker 00

  Yes, I do think this is different from crypto. I think this has a bigger impact on society than crypto will. Because this has the type of impact that electricity and the steam engine had in terms of productivity changes.

• 35:17Harry Stebbings?

  So you don't, so that's fascinating. So you don't think that, Barney, the funding hype is overhyped? You think this is justified? Do you think this is a fundamental platform shift like mobile wars?

• 35:26Speaker 00

  Yeah, of course, the funding hype is overhyped. It always is. But it's also because a lot of money has to go into it and we have to try a lot of different things to figure out which ones work. So I do really think that AI can change where we go in terms of how big companies need to be to do things and how many things we can do on a day. Using machine learning to do so much of the work that people are doing on a day-to-day basis could be very powerful. But we need to get better at actually understanding what is machine learned. So we need explainable AI in a way. We need to actually understand what it's doing. We can train a model to say, hey, you are getting credit, you are not getting credit. But in the end, we actually have to be able to explain why it's making that decision.

• 36:11Harry Stebbings?

  You mentioned how it changes development cycles. Currently, I think it was GitHub posted that forty-one percent of code today is AI-generated. Five years' time, Yoast, how much is then?

• 36:21Speaker 00

  Sixty to eighty. It allows you to focus on business logic and a whole lot more.

• 36:25Harry Stebbings?

  Unpack that. What does that mean?

• 36:27Speaker 00

  The hardest part right now of building software is thinking about DUX and then fixing the business logic. But the business logic is almost always like only ten percent of the time you spend on this whole thing. Everything else is all the other stuff. And all the other stuff is relatively basic, but it just costs a lot of time. But the relatively basic stuff AI can do fairly quickly. So you need to create the processes, you need to create the logic, because the AI is not going to do that for you entirely. But at the same time, writing the The individual functions is going to be easier and easier. I'm a big fan of CloudFlare. If I describe to OpenAI, like, I need a CloudFlare worker that does this, this and this. It'll just give me the code.

• 37:11Speaker 00

  I don't have to code anything anymore. It'll just give me a working worker. I've done this twice in the last few weeks where it just did exactly what I needed it to do. I didn't even have to write a single bit of code. And I think that's going to be a lot more common. Think of an OpenAI interface on top of no-code interfaces, et cetera. Connecting APIs together is going to be tremendously powerful and it's going to save a whole lot of time coding things.

• 37:35Harry Stebbings?

  Final one before we do a quick fire, because I could talk to you all day. When you decide, hey, I'm going to angel invest, do you set aside a budget? Do you decide I'm going to write 25K checks, 50K checks? I'm always going to write the same check. When you moved into angel investing more and advising more, how did you think about the strategy there?

• 37:53Speaker 00

  Not enough. I tend to write slightly bigger checks. The thing is, I want a company to be able to be self-sustaining as fast as possible. People send me decks with like, yeah, we were doing an angel round now and then we're going to do an A round in a year. And then they've already planned like when they'll do their rounds, but they have not planned when they'll build the product. I sort of glance and go like, no, I'm not interested.

• 38:16Harry Stebbings?

  So you guys write like 250K checks?

• 38:19Speaker 00

  200, yeah, 100, 200K is usually, yeah.

• 38:22Harry Stebbings?

  Okay, how many do you want to do?

• 38:23Speaker 00

  Probably another 20 to 40 in the next five years.

• 38:28Harry Stebbings?

  How involved are you?

• 38:29Speaker 00

  Very. So we have a shared Slack channel with each and every one of them. And I talk to almost every one of them on a, not always daily, but most of them on a weekly basis.

• 38:37Harry Stebbings?

  Do you love angel investing?

• 38:39Speaker 00

  I really, really love working with founders and seeing their energy and being able to help them grow.

• 38:45Harry Stebbings?

  Do you mind where you invest? Does it have to be pre-seed? Can it be a series A, B round?

• 38:49Speaker 00

  I think we are both, Marieke and I, are better in the pre-seed, seed stage than much later on. If you're getting to later on, then most of the value that we can probably give you has already been happening.

• 39:02Harry Stebbings?

  Do you find most of your deals are in Europe?

• 39:04Speaker 00

  No, I don't care.

• 39:05Harry Stebbings?

  I want to move into a quick fire, Yoast. I say a short statement, you give me your thoughts. Does that sound okay?

• 39:10Speaker 00

  Yes, absolutely.

• 39:11Harry Stebbings?

  Okay, so what did you believe about growth five years ago that you no longer believe?

• 39:15Speaker 00

  That it was easy. Also, a bit of an underestimation of what Marika and I could do together.

• 39:21Harry Stebbings?

  Tell me, work-life balance. What do you advise me, Yoast? We spoke about it before.

• 39:26Speaker 00

  Move out of London. I think work-life balance is absolutely achievable. Also, don't think it means what people think it means. If it's good for you, then it works, right? I will say that not being in a big city and having to go to a fancy dinner three times a week does really help.

• 39:42Harry Stebbings?

  Can a startup be built on free traffic as such in 2023?

• 39:46Speaker 00

  Yes.

• 39:46Harry Stebbings?

  Single biggest mistake people make when starting SEO strategies today?

• 39:50Speaker 00

  Not considering branding from the get-go in their entire strategy and not considering what words people use to describe their problems.

• 39:59Harry Stebbings?

  What would you most like to change about the world of growth?

• 40:01Speaker 00

  I would want people to stop thinking they can measure everything, and that's mostly stolen from my wife, who's a PhD in sociology and just says, people think that they can explain behaviour way too much, and you can't.

• 40:13Harry Stebbings?

  Final one. What one company growth strategy have you really been impressed by recently and why?

• 40:18Speaker 00

  I think that what Microsoft's doing with Bing right now is absolutely brilliant. I really love it. So Bing Chat, it's very fun what they're doing there. It's a bit cheeky because it doesn't really work yet, but it's making everybody at Google so very conscious of what they don't have. So I really like that. But in terms of other companies that I really like, I'm always impressed when companies do marketing towards developers very well and grow on top of that. Because I do think that that's still an underappreciated way of going at the market. So companies like Algolia that actually do great developer outreach, provide absolutely awesome docs on how to work with this software and how to solve your problems with it.

• 40:58Harry Stebbings?

  What makes awesome doc?

• 40:59Speaker 00

  Well, just good documentation of everything you can do with your software and also copy-paste examples that actually work.

• 41:07Harry Stebbings?

  We mentioned great tech platforms earlier. Webflow, good, not good?

• 41:11Speaker 00

  It competes directly with WordPress, so I'm always going to say not good.

• 41:14Harry Stebbings?

  Squarespace?

• 41:15Speaker 00

  Reasonably good.

• 41:17Harry Stebbings?

  Wix?

• 41:18Speaker 00

  Good growth tactics at the moment. They're really focusing on the SEO industry as one of their levers of getting other people to install or use Wix. I don't really like their product.

• 41:28Harry Stebbings?

  Yo, listen, I cannot wait to do many more angel investments together. Thank you so much for putting up with my prying questions. I've got a lot more prying over the years. I used to be very easy as an interviewer, but you've been a star, so thank you so much. Thank you for having me. It's been an absolute pleasure.

• 41:45Harry Stebbings?

  What a fantastic guest. I so enjoyed that show. If you want to see more from us, you can head over to YouTube and watch the full interview by searching for 20VC. But before we leave you today, I need your help. I need your input on the 20VC Miro board as I'm booking new guests that we should feature on the show in 2023. Now, this is the only way to make suggestions. Head on over to miro.com/20vc and leave your suggestions with digital sticky notes or comments. That's miro.com/20vc and Miro actually sponsored this episode. For those of you that have not used it before, it's this incredible tool that I think is truly game changing. Just ask the 1 million plus people who sign up for Miro every month. It's a visual collaboration platform packed with the right tools, tech and templates

• 42:30Harry Stebbings?

  to grow your idea, your product, your vision at scale. Miro acts as a single source of truth across teams. That means you can get buy-in much faster from teams like marketing, engineering, sales, product, basically anyone who needs to have a say to make growth projects grow. And even though it seems like a simple digital whiteboard, it's so much more. Cashing out a sound acquisition strategy, brainstorming Miro, rethinking your sign-up flow, using Miro's user research templates, quickly visualize team structures on a Miro board. However you need to grow, Miro can help. We use Miro all the time to brainstorm future shows, vote on potential guests and leave notes for the rest of the team in our own time.

• 43:10Harry Stebbings?

  Check it out again at miro.com/20vc to submit your guest suggestions. And I'm really excited about this next sponsor because a few of my portfolio companies use it and love it. So do you know the impact of your marketing efforts? It's more important this year than ever to know the impact. Or how much you've spent so far this month and what you got as a result? You're not alone in not knowing these answers. Even though marketing is now all about the data, most companies still use manual spreadsheets for reporting. An analysis, but if you want to supercharge growth, spreadsheets just won't cut it. Funnel is the world's first and only marketing data hub that takes all the manual effort out of data collection and lets you blend data from different platforms to unlock new, deeper insights.

• 43:54Harry Stebbings?

  And the best part is this can all be achieved without writing a single line of code. I can hear your sigh of relief. That's why it's used by more than 2,000 of the world's top brands and media agencies, including Samsung, Adidas, Sony and Havas Media. Start your own free trial by visiting funnel.io/20vc. That's funnel.io/20vc. And finally, we've all been there, but nothing is quite so depleting as the blinking cursor on an empty page. Reminds me of law school. You're sitting there trying to come up with ad copy or hypothetically a set of interview questions for a podcast you run, obviously personal, and it's just staring back at you like a brick wall. This moment of frustration amid growing content demands is probably why I'm hearing about Jasper non-stop these days.

• 44:39Harry Stebbings?

  Jasper is an AI assistant for business that helps marketers, sales reps and other creators accelerate their writing process and repackage content into different formats, languages and tones. Unlike broad use AI tools, Jasper was built for business and this makes a big difference. It incorporates SEO, a plagiarism checker and a browser extension. So you can call up Jasper in pretty much any text box or platform you use online. Visit jasper.ai/20vc for a demo. Your burnt out team will thank you. As always, I so appreciate your support and I can't wait to bring you a fantastic episode this coming Friday with David Marcus.