Show-don't-tell Launches & Fighting to (not) Persuade

But the idea of, I do bolt.new and it just works. And to be convinced, at bolt.new I actually get GIFs on what I'm gonna get and how it's gonna work. That's the ultimate show, don't tell. There's no landing page, just go into the product. And I think more and more that for developers, if you can do that, if you can get them to the product and leave that wow experience fast, I think it's fantastic.

The moment to wow. Any concept like this where you start measuring your team on getting there faster. And also the .new domain is fun and there are stipulations on it. It has to initiate something. You can't use it as a marketing page. It has to actually activate something.

Hi everybody, my name is Martin Gontovnikas, but everybody calls me Gonto. I'm here with Hank Taylor. Every week, Hank and I boil down the most important product growth and marketing learnings from what's happening in DevTools. So in simple terms, we give you our thoughts on the current thing in the DevTools space. We hope you enjoy it.

Here we are, episode two. We've seen a few things happen. Been trading notes, you've been on vacation. We had a StackBlitz launch of bolt.new that I wanted to talk about. First, StackBlitz helps run ViteConf, which in itself is already a clever partnership. StackBlitz does a lot of the work, Vite gets a lot of benefit there, but then they get this massive distribution engine and it's all open source friendly, yada yada. But then Eric, the CEO of StackBlitz, gets to do the keynote or one of the keynotes. And so this year they introduced — it's effectively a v0/Replit competitor — bolt.new. And there were a couple of cool things about the launch. I mean, one, it was smart to launch it at the conference. So instead of a standalone product launch, they have this huge registered audience that's watching it and coming for all the other talks. So also shipping at a conference, they've got all the captive audience that's there for all the other speakers and other reasons. And shipping something at a conference in a big public way with a big public date, you and I both know it forces the developers to work really hard. In fact, I actually DM'd Eric about it and he told me he hasn't slept for like 72 hours leading up to the thing. I'm sure squashing last-minute bugs. And yeah, the tweets — there were some things I think you liked about the tweets.

First of all, I really liked the idea of conference-driven development. When I worked at Auth0, we actually had two years that we weren't shipping anything because we had contractual commitments and stuff like that. And what got us back to shipping was actually doing a conference and pushing to do a conference to ship stuff for there. I know Vercel has done it in the past, so big believer on that. What I personally liked about the tweets is that I see more and more companies — but not all doing it yet — that they use GIFs to show the product. So if you look at bolt.new's tweet, you'll see that they have a thread of multiple tweets, one below the other, where in each tweet they have a GIF where they show some functionality and how it works and why it's awesome. I'm personally a big believer on this idea of show, don't tell. I talk about it a lot with all of the companies that I work with. And if I think about developers, developers have this high bullshit sniffer, as I call it. So for developers, if you tell them, "Oh, it's fantastic, it's great," and stuff like that, they will not believe any of that. So with developers, I think there's two things that are really important. One is show, don't tell. Every time you have a chance to actually show how the product works, show how it's great, just do that instead of telling people why it's fantastic. And secondly, I hear a lot of product marketers always talk about this idea that you should talk about value propositions and what's the benefit that you get and stuff like that. I don't think that's true for developers. I think what developers love is specificity. So being very specific on how does this functionality work and what does it do for you. And I think that the GIFs and the show, don't tell actually do that for you. Do you agree with that?

Yeah, because developers more than most other personas, you know, at least in our world, they wanna see the tools and they wanna figure out, how does that fit into their world, more than anything. The show, don't tell is — if you put a hammer and a nail on their desk, they're gonna start looking at them and go, "Oh hey, bang!" They wanna get to that. And I'll add, part of the reason I love this launch is because when you type in bolt.new, it jumps you right in and you can type a prompt and it immediately just starts doing things. And I was delighted at how quickly it did that. And then I started reprompting it and whatever. And it was a really delightful experience.

And I think that in that case, it's literally the most show, don't tell. They don't even have a landing page. Like, people who have a landing page, I tell them to have something interactive. But the idea of, I do bolt.new and it just works. And to be convinced, at bolt.new I actually get GIFs on what I'm gonna get and how it's gonna work. That's the ultimate show, don't tell. There's no landing page, just go into the product. And I think more and more that for developers, if you can do that — instead of just even having a landing page that shows — if you can get them to the product and leave that wow experience fast, I think it's fantastic.

Yeah, the moment to wow is really short. At Vercel, we used to have a "time to first confetti," which is the first time you get a deploy. And like any concept like this, where you start measuring your team on getting there faster. And also the .new domain is fun and there are stipulations on it. It has to initiate something. You can't use it as a marketing page. It has to actually activate something. So I liked it, I thought it was a cool launch. There were other launches that I told you about, like Builder.io — they announced one where they had a similar animated GIF on the page and they had a fun little ticket customizer. I love the tickets. Yeah, I like the trend, but like you said, not enough people are doing it yet. I think more people need to do show, not tell.

And to me what's fascinating is, I see more companies now starting to do launches because Supabase, I think, started — like Supabase was doing these launch weeks, where in a week they launch every day one thing. And I've seen literally every company do something like that, but not everybody's doing it right. Because number one, I think what Supabase does great is they do a lot of this show, don't tell where you actually see how it works. They give you a video, a GIF, a demo, and they literally show you. And number two, what I think Supabase does great is that most of the features that they ship on the launch week, they actually partner with somebody else. So they do a feature with somebody, and in those cases they get this value where each feature will be released with the audience and the users from the other partner. So they get to talk not just to their users but also to others. I've seen a lot of companies do launch weeks where they just ship features, they don't show them as much, they just tell you, and they don't do it with partners. And then they expect everybody to know about it, but there's no viral component, there's nothing to show, there's nothing to use, there's no new partner audience. It makes no sense.

Yeah, a topic for another time I think is like surprising people with features versus doing the collaboration and stuff, because both can work. They're extreme opposites but they have different use cases. So speaking of showing and not telling and using visuals and whatever, and proof to persuade, I think you had a great topic on this.

One of the things that to me is fascinating about the front-end world, and the development world in general — I think mostly in front-end more than anything else — is this thing where there's so much drama and so many arguments online. As a lover of gossip, I love them personally because it's fun. But my question is, is it useful for the companies? I've seen two things happen in the last two weeks. Number one was this whole thing in web components, where there was a tweet that somebody said that web components didn't make any sense at all and that they didn't help the web. Then I saw Igor Minar, who used to be on the Angular team, saying that they were going to do web components but in the end decided not to. And then Rob Eisenberg, who is actually the creator of web components, came in and started to discuss with Igor, like, "No, you didn't really help us out and you didn't come with an open mind and it wasn't the right thing." It was this big drama that everybody was talking about. Same week, there was another big drama, I think, between Vercel and Cloudflare, where Vercel started to talk a lot about why CDNs are a thing of the past and then started to talk about why deploying Next.js in a place like Cloudflare isn't as good as on Vercel, and started to give specific information and specific things about it in a tweet where they actually mentioned Cloudflare. What's your take on these online dramas and arguments? Are they useful for the company, useful for the community, useful for who, and how do you think about them?

I think it's usually — if it's useful for anybody, it's useful for the underdog, for the person introducing that like, "Hey, the incumbents have it wrong, or the way you do things, reconsider it," because it might get some people to do that. But when you're becoming a market leader, I don't think it's persuasive, first off, to go and argue with the underdogs. And notably, I don't think we've seen Cloudflare arguing with Guillermo Rauch on this, but we have seen other people arguing with Rauch on this, including partners of Cloudflare who are smaller, and therefore I think they're the ones benefiting from it. And yeah, I think engaging in the drama directly or stirring it up that way with these kind of ivory tower arguments — it's hard as an outsider to understand: okay, is this a good faith argument? What are they leaving out? Do they know they're leaving out stuff? Are they really presenting the best case here? Like, it's Twitter, so probably not. And they're heavily incentivized to probably not make good faith arguments online. Like, none of us are when we just want to be right and persuade people. I would advise against it almost every time.

I think that the company for whom this is useful is other companies that are not in the battle. What I mean by that is, every time I think about doing a content strategy for any company, I think a category of things that you should do is what I call "surf the wave." And surf the wave is: look at what's the current thing on Twitter or in the space in general, and then try to write content very fast about it to see if it becomes viral and if it catches on. So that's what I would do. I would actually not talk bad about the companies, but rather surf the wave if it's related to me, to get people to click on me and see it. For example, in this Cloudflare versus Vercel thing, one thing you could do is write an article comparing how using a Next.js website is different on a CDN versus Vercel, or thinking about showing the difference between a CDN static website compared to something else. So I think diving into the problem, but without judging the companies and actually just showing data or talking a bit about it, could be a way that you get content to go viral and then you show up more. Of course, this is for the underdogs. This is for smaller companies that want to hijack the trend and then start talking about it.

It can be a double-edged sword, because to some of these people it might appear you're taking a side. And something I'm always telling people is, the best thing you can do in any sort of marketing — but I think it's especially in devtool marketing — if you can get whatever your thing is to be somehow tied to the developer's ego, identity, whatever, then you're gonna win. Because we don't identify with companies as much as we used to. Everybody job-hops a lot. But we do with tools. We'll use the same tools. We'll have tools as requirements. "You must use this stack for me to even apply to work for you" type of stuff. And so devtools are becoming part of people's identity. So I don't know if there's a way for you to go experiment, create data, surf the wave, and do it in a way that helps reinforce identity. I think it's just a great thing to do, though there will be some danger. Sometimes the wave will crash on you or something.

There's always risk, but with no risk, no reward. If you're an underdog, you can do this because you have very little to lose. That's why the big companies would never do it. At the same time, I think that in general, for any company, whoever they are, it's always bad PR to talk bad about other companies, because talking bad about somebody else will always generate hate. And I think you can talk about it without mentioning the other company. You can always spin it and potentially do a piece of content or a tweet or something where you talk about the great things about your product that the other product doesn't have, without mentioning the other product. So you say, "Oh, my product has this, this, and that, and that allows companies to do this thing, this thing, this thing that nobody else can do." So taking a positive approach, talking about differentiation from others but from a point of view of, "I'm proud of what we built and this is why it's great," and not mentioning competitors — I think it's the best way to tackle this. In general, talking bad about somebody else, I think, always has a bad taste.

Yeah, and even in this example, I could think of how maybe Rauch could have done a tweet about why Vercel isn't building a CDN like this. He kind of didn't even necessarily have to mention Cloudflare, or let alone directly reply to the underdogs that were effectively baiting him into conversation and bringing his many followers to smaller circles.

What's interesting to me is that I don't see — and I don't know why yet — but it happens that in the front-end world, people, as you said, relate their identity to a platform and a framework and stuff like that. I don't see that happening as much in the backend community, or SRE, or DevOps communities. So thinking about that, I think, is important when you think about your marketing and how to go after some of these frameworks and technologies.

It's still definitely there, though. I mean, there are infra, DevOps folks where Kubernetes is life — or "I hate Kubernetes" — and you get all sorts, just as one example. But you see it in everything. I replied to a tweet that you sent me related to this and I pointed out that it happens with Salesforce. And I've had employees make terrible decisions because they were so identity-driven around Salesforce, which I think some developers listening to this would laugh — that people identify or tie some of their identity to Salesforce — but we all do the same thing.

So if you have any feedback, feel free to tweet either to Hank or to me with any feedback on this. Thank you!