Tanay Pant is an author, speaker, developer and the head of developer relations at n8n.io, a workflow automation company. He has written three books, “Learning Web-Based Virtual Reality,” “Building a Virtual Assistant for Raspberry Pi,” and “Learning Firefox OS Application Development.”
In this Hoss Hangout, we talk to Tanay about the power of great content, including its ability to shape the culture of a developer community, and how to approach creating a meaningful content strategy from scratch. He also shares wisdom about collecting important feedback from a community and lessons learned from his experience as employee number three at n8n.
- Tanay talks about how content strategy - and even word choice - can shape the culture of a developer community [3:51]
- Tanay walks us through how his team collects feedback from the n8n community and translates it into valuable content and product updates [6:30]
- Tanay explains that there is a key difference between audience and community [11:33]
- Tanay talks about starting and growing n8n’s content strategy and shares some tactical tips for others in his shoes [15:37]
- Tanay says it’s important to view your documentation as a product and conduct usability testing on it [22:03]
- “Much like companies and products, developer communities are unique, as well. And, while there might be an overlap between different communities, there usually isn’t a one-size-fits-all.”
- “Treat the documentation like a product. We’ve got to accept the fact that it’s a living thing, it needs attention and will never be finished.”
- “Do what works best for the community and don’t be afraid to try out something radically different, because it takes time to get insights into what is working and isn't.”
Matt Hawkins: So, we’re excited to be hanging out with Tanay Pant today. Tanay is an author, speaker and developer. He is the head of developer relations at n8n.io and has written three books, “Learning Web-Based Virtual Reality,” “Building a Virtual Assistant for Raspberry Pi,” and “Learning Firefox OS Application Development.” Tanay, it’s a pleasure to have you.
Tanay Pant: Hey Matt, thanks for having me today.
Matt: Yeah, absolutely. Well, we have a lot to cover. I’d like to maybe start off diving into content and creating a great developer experience. I know this is something we’ve chatted about in the past, but just to kick things off, how important is content to creating a great developer experience and maybe as a follow-up to that, what role does it play in other aspects of developer community or otherwise?
Tanay: I think content is one of the essential pieces of the puzzle when it comes to creating a great developer experience. From a developer relations perspective, it’s the base of a lot of activities that we do. Content can help us build awareness about a product, help people get onboarded and inspire them to build something that makes an impact in their lives. Not only that, it also helps nurture the community. So, at n8n, we have three loosely defined focuses for the content we create. So, the first kind would be onboarding. We like to say that this helps solve the blank canvas problem. So, essentially, people discover your tool, they find that it’s cool, but oftentimes, the question that comes is, “okay, that’s all nice and good, but what do I do next?” So, that’s the kind of content that handles onboarding. Then, we have education. So, this talks about what powerful things you can do with certain tools or platforms, but it also covers the more conceptual topics. And finally, we also do community interviews. So, these are stories that are shared by the community about how they use n8n. And we feel that the stories that are shared by the community are often the most powerful ones to inspire other community members as well. So, these are like the three loosely defined focuses that we have right now from a content perspective, but of course, I expect the strategy and the focus areas to evolve as the content base matures as well.
Matt: I’m curious, which of those three areas is or was the most challenging to start with?
Tanay: So, we started off with onboarding because from my personal experience, like, I discovered n8n, I really liked the tool, and n8n literally has a blank canvas when you open it. Like, there’s the editor UI where you can add notes, where you can create your workflow, but it starts with an empty screen, and that was an interesting topic for me to think about is like, okay, it has a lot of integrations, and a lot of them are tools I use, but how can I actually connect these different apps or integrations to create something meaningful? And I was new in the automation space, so I didn’t have a lot of ideas already as other folks might have had. So that was, I think, something that was quite challenging, because we have a very diverse community, people are doing a lot of different things with n8n. Like, people are using it for IoT, they are running it on Ships, they are using it to sync their CRMs, like a lot of different things that are happening. So, I think that was initially the most challenging, but something that became the most fun over time.
Matt: Absolutely. Critical component of a great content strategy. So, jumping to community. You know, you’ve said in the past that content can play a really big role in developing a community, but also, in addition to that, shaping the culture of a community. I’m curious kind of what you mean by that and if you could kind of expand on your thoughts there.
Tanay: Sure thing. So, at n8n, we focus on creating an inclusive environment for the community. So broadly speaking, that means rude behavior is not tolerated and every question is treated with the same degree of respect. So, we encourage people to ask questions, no matter how elementary they think that their question might be. After all, if people are having issues and they’re asking basic questions, that could mean either our product’s usability can be improved or that we can provide better content or documentation for education. So, by creating content, we make sure that our tone of writing reflects our approach to community management. So for instance like, “simply,” “easily,” “just” - they don’t make it past our editorial review. And, this might sound like a practice that isn’t very significant, but it can really make all the difference. Now, I didn’t know that when I was starting out with my writing career. So, as I interacted with more companies and their content, I realized how serious this can be. I remember I discovered a product that used these terms, so I’m going through their content, maybe a tutorial, and it would read like, “setting this up is very easy.” And I’d think, well, this doesn’t seem that easy to me. Maybe this isn’t the right product fit for me, right? And that’s a great way to lose people. So, more than that, what happens is the community adapts to things like this. So, next time somebody asks a question in the community, the folks will refer to that tutorial or docs and reply, “well, it’s quite an easy thing to do.” Which might dissuade people from then participating in the community or from using the product. So, having looked at this, I feel like it’s often the small things that can make the biggest difference in this space.
Matt: Yeah, shaping that dialogue and shaping the attitudes in the community. Yeah, that’s fascinating. So, you mentioned every question is respected, every question is encouraged. What does that feedback look like? Or that feedback loop look like? With your team, you know, you’re in the community, you’re working with your own product team, I’m just curious like tactically what that looks like in terms of taking those insights and making sure that, you know, the team acts on those.
Tanay: I think feedback is really important from the perspective of what we are doing in the developer relations team. Like, we really keep an eye on this kind of feedback because it depends on that, and that correlates to basically what our content strategy would look like, because it doesn’t do anyone much good if we create content that nobody wants. Especially with 2020, a lot of online events are canceled, everyone is creating content, there’s a lot of content on the internet, right? So, what we do is we synergize between different functions in the company to find out what is most valuable for our community. So, tactically, our DevRel team is embedded into the community, so, we answer questions, we try to understand the pain points, and over that, we try to recognize the different trends that are coming through. More often than not, we see common threads around certain topics, which makes it clear around what topics we need content on. So, if we already have content on it, it gives us insights into what needs to be better. We also do regular chats with our users, take part in user onboarding to give us a deeper understanding of what is the most valuable type of things for our community. And honestly, this is nothing that developed overnight. It took us quite some time to recognize what works best and evolve our way of operating based on these learnings.
Matt: Yeah. So if you don’t mind, one additional follow-up question there. Just super tactically, I’m curious, because I think this is something that a lot of teams struggle with, you know, you and your team are in the community, you’re hearing various things, various opportunities. Is that something where just you get a sense as a team kind of what those big themes are? Or is it something that you record, I don’t know, in a spreadsheet or a Trello board like every complaint or every, you know, feature request, and then from there you kind of just keep a tally of how often those are mentioned?
Matt: Yeah, I think a lot of teams sort of under-appreciate how much you can learn from your community. You know, I think a lot of teams, when they think about building, launching and maintaining a community, they think more about, whether it’s a support or lead-gen perspective, but yeah, just the insights you can garner can be incredibly powerful. So, one final question on communities. What are the, you know, we talk to a lot of companies who are just getting started with their plans to launch a community. What are those first couple of things that companies should do to go about building a great community?
Tanay: I feel like, sadly, there isn’t a magical scroll that describes that, as much as we’d love that. But, much like companies and products, developer communities are unique as well. And, while there might be an overlap between different communities, there usually isn’t a one-size-fits-all. Often I’ve seen companies make the mistake of going all-in to build a community, and if you ask them why they’re doing that, you’d get the answer that they feel that they should. Probably because of peer pressure. That’s a new common thing in the startup landscape these days.
Matt: For sure.
Tanay: Or maybe because they want to convert them into leads just like you mentioned. But, you can’t really build communities with such a transactional agenda or without knowing why you’re doing that. So like, audience and community are really different things, and what some folks mistake is building an audience is not the same as building a community. So, let me clarify that by means of an example. So for instance, we have a community-backed product at n8n. The initial community was comprised of fans who tried out the product, found out that it created value for them, and decided that they wanted to hang out to see what’s next. And we wanted to create a place where people can discuss automation, ask questions about n8n and get value out of. Now, I realize that might sound very idealistic, but another example of that is we are very active on Twitter. So, we don’t follow any accounts other than that of Fair Code. So, we want our community to follow our Twitter account only if they feel that they get some value out of following us. Because people often follow you back if you follow them. And the number of followers is really just a vanity metric for us. So, we don’t really do traditional marketing on our social media channels, but try to provide high-quality content and share things that might be helpful for them. Because if you operate in this fashion, you build both the right audience and the community, and I feel that a lot of times, content is not good or bad, it’s probably just relevant for some and irrelevant for others. And you know, you interact with the community, you can really get a sense of the company’s philosophy behind it. And, people know.
Matt: Yeah. People know. So, curious, one follow-up question, so you mentioned, you know, you share a lot of content on social, what type of content generally performs the best with your community? And I guess we can just define that as like, you know, positive interactions or positive engagements. Just curious, is it like video or blog posts, guides?
Tanay: Sure, so we didn’t really have a formal social media strategy until December, or, let’s say late November. Before that, what we used to do is pretty much share major updates. If we had some new releases, you know, we release every week, so there’s going to be a post about that, but mostly if we write any content or we create something new, we pretty much just cross-posted it across the different social media assets. And by content, this is mostly something that is either an educational source, onboarding, like of the categories we talked about. So, for us, what we saw perform the best were either Tweet stories, so these are something that we started experimenting with late last year, it worked out quite well. So, we go ahead and we talk about, you know, we were trying to solve a problem, and this is not a hypothetical problem, this is something that we actually faced on a day-to-day operational basis, and then we explained like, this happened, and then what kind of approach we took to solve that. How we used n8n to actually solve that problem, and what’s the current state. And often, we’d go back to this sort of content and post updates like oh, we did this, now we found out even a better way of doing this with n8n, this is how we updated. So, that kind of content gains the most engagement. After that, I think these community interviews we are doing. So, we have a chat with the folks using n8n. Like, as I mentioned, being a community-backed product gives n8n incredible, amazing diversity in terms of what people are using it for. So, you know, we often have stories of people who are using it to run n8n on ships. That’s amazing, like, never would have thought of that. Then, we also have people doing IoT stuff with it. So, these stories by the community also do really, really well.
Matt: So, taking a step back for a second. I know you were employee number three at n8n and you started all of the content yourself, right? You started the blog, and you know, the community and everything. I’m just curious, like, what was that like? And do you have any recommendations for individuals that are in your shoes, or were in your shoes when you started all this just by yourself? Because it’s a big feat for a big team to take, it’s an enormous feat for an individual.
Tanay: Yeah, oh man, that was so much fun. Well, I mean, I didn’t exactly start from scratch. We had great documentation on the core part of the product. That was really helpful, and a great base to build on top of. I remember the docs were on Docsify. We did an audit of the popular documentation tools, and because of various reasons, we decided to migrate the whole thing to View Press. We started with building a Medium publication for sharing content on n8n. So, we didn’t have a very spearheaded content strategy at that point because we didn’t have any content, so we wanted to solve our blank canvas problem that we chatted about earlier. We started off with Medium because we thought that building our own blog would take too much time. Especially with the approach that we had in mind at the time. And we wanted to move fast. So, I started with the approach of noting down the pain points that I observed relating to the product. At times, I had to ping my team members to get an explanation on something. And these instances really helped me come up with the initial content. So, if I had doubts or questions as a new user like the others might, too. So, I think that’s also a great way of doing employee onboarding, as well. Especially for the teams focusing on developer experience because you really see a product with fresh eyes only once, and that way when a new colleague is learning something during onboarding, they’re also creating value for the community.
Tanay: Yeah. And, after that we started cross-posting on Dev.to and Hackernoon, so that worked out really well. And, looking at the last 90 days, we have had about 22,000 minutes of content read. Over time, we realized that we were taking SEO hits by not having the blog on our domain. Medium also had other limitations in terms of, you know, number of authors you can add to a post, statistics, among other things. So, right now we are in the process of doing a migration away from Medium. But yeah, like, I can recommend cross-posting to Medium. So, looking back, this is something we should have done differently from the start. But yeah, good learning.
Matt: No, I’m with you on that. I’ve never been a huge fan of Medium for that reason. I think you’ve got to own your own content, you’ve got to get the SEO benefits from that. I think it’s wise that you’re thinking about Medium more as a syndication channel similar to a Dev.to, than kind of the origin of the content itself. But yeah, when you’re scrappy and you’re just getting started and you don’t have time to spin up a blog, then, yeah, it’s obviously better to do that than not produce any content as well.
Dev.to, Hackernoon - for syndication, if you had to just pick one, what performs the best?
Tanay: Hackernoon, for sure. But you can easily run into challenges with Hackernoon. Because I think it needs to be really developer-focused content that performs well, which was fine because our content was really educational rather than marketing-focused. Dev.to, I’ve seen it work really well for some companies, so it really depends on how much of an audience you already have. Maybe from your personal accounts or your company account. So, if you already have something, it’s really easy to sort of cross-post content there and get some good engagement on that. Building it from scratch in Dev.to is a bit more work than the other channels.
Matt: In terms of getting into, you know, kind of a cadence, teams start producing content, they’re publishing on their blog, they’re syndicating - do you have kind of a good recommendation on how often they should be syndicating out? Is it something that once a week is fine? Should they be doing it daily? Like, if you’re starting from zero and you want to build up your audience.
Tanay: I think daily content gets a bit too much if it’s just one person working on content, I think daily might be a bit too much. Because like, I’ve seen that in some companies where individuals have to create new content every day, and especially in a role like developer experience or developer relations, you’ve got to do a lot of learning, because, especially as part of early-stage companies, product matures really quickly, new things get added, you’ve got to learn them. You’ve also got to read a lot of content to get some good inspiration of things to do. I think like daily content, if it’s just one person handling, is unfortunately a quick way to burn out. But I would say, it should be done in a way that works best. I think it’s quite okay to start off once a week, because, it’s like any other thing. You get started with it, you get better at it and quicker at it. I remember I used to write every week. Lately, I’ve been a bit lax in terms of writing and starting a new content piece is going to take me much longer than how I was writing earlier.
Matt: Yeah. When you’re not into that rhythm, and yeah, not sacrificing on quality, either. Starting out with, you know, maintaining high quality once a week and yeah, for sure. You mentioned a few moments ago about documentation and kind of your early docs there. We all obviously know docs are the foundation, it’s the table stakes of delivering a great experience. I’m just curious if you have any tactical tips for teams when it comes to creating really great docs. It seems like almost nobody is happy with their own docs or other people’s. So, I’m just curious kind of your thoughts there.
Tanay: Well, I think a lot of companies have great documentation. But I think, this is really important, and this is something we have been talking about in the company as well, is to treat the documentation like a product. Like, we’ve got to accept the fact that it’s a living thing, it needs attention and will never be finished. So yeah, just like a product, you’ve got to do usability testing. Because this can really bring out shortcomings in your docs that you, as an expert, might have never thought of. And, we did the same thing for parts of our documentation which are extremely important like quick start or creating your first workflow in n8n. And the insights that we got out of these sessions were immensely valuable. Just as an example, so, one of our creating your first workflows back then was, you know, it’s a workflow that runs every day at 8 a.m., gets the weather information for your city, if it’s less than 16 degrees, sends you an SMS, “hey, take a sweater with you.” Or, you know, if it’s not then it doesn’t. So, we made it incredibly detailed and we thought, you know, it’s a great content piece, people will find it easy to get started with n8n. So, I think we learned about this practice of doing usability testing from the folks at Twilio. So, I thought, okay, let’s do this, it’s going to be an interesting learning. So, I got one of my friends who had never tried out n8n. I was like, go through this page of the documentation, think out loud, I’m going to put myself on mute and I’m just going to observe you. So, it was funny, he started off with the crown node and he added the time, okay, run at 8 a.m. every day. He tries out the workflow, doesn’t work for him. He’s thinking like, why isn’t that happening? Because the default time zone is New York and not Hamberg or like Berlin, CSE. It’s like, that’s a thing that we never thought about. Going on then, we see smaller things like, there’s a phone number field in the Twilio node. The person goes on zero, ABC, entering their phone number. And it’s like no, no, you’ve got to put the country code there. Okay, these things we never thought about as the people who wrote the documentation use the product so much regularly. So, that was amazing to gain those insights and fix these things.
Matt: You’re so right. Sometimes you’re just too close to the content, right? You’re living and breathing it, which is great, but, at the same time, having that fresh pair of eyes and somebody like a close friend or just bringing in your users and watch how they experience with brand new eyes can reveal some of those small things that you just kind of took for granted.
Tanay: Exactly, yeah. And I remember I started writing nodes for n8n recently, and nodes are closely modeled after a different API, so I was working on a node for Google Translate. And, you know, you have credentials, you enter the text that you want to be translated - Google, in its API, calls it query - and then, you know, the language you want to translate into. So I was like okay, query, I called the field query, and just sent in a PR. So, thankfully, we have usability reviews before actually merging in the nodes. So, somebody was doing a usability review and they were like, what’s query? Like, do we have to have some kind of boolean logic or is it like sequel or what is it? No, no, it’s just the text you want to translate. It’s like, shouldn’t it be called “text”? I was like, you’re absolutely right, it should be called “text.” I never thought of it because I was copying the API and then like okay, query, query.
Matt: Yeah, and move on, you’re moving fast, yeah absolutely. That’s funny. Well, you know, hey, this has been really fantastic, Tanay. Thank you for sharing all of your best practices and insights. Got one final question for you - or two final questions. Do you have any parting thoughts, words of wisdom for other teams who, they might be starting this DX function or DevRel or looking to kind of reinvent their program. And also, where can people find you on the web?
Tanay: Sure. So, I’d say, like, we talked about a couple of different approaches, but I feel like, having seen this for a while, there is no one correct way of doing this. Like, do what works best for the community, don’t be afraid to try out something radically different because it takes time to get insights into what is working and isn’t. Also, a lot of times, you’d find that your community and the community for some other products have a degree of overlap. Perhaps you’re both in the no code space or both are focusing on DevOps. So, don’t underestimate the value of such collaborations because such partnerships can give rise to content that’s very interesting and helpful for both the communities, so that way your content will help you reach out to folks who might not have heard of your product before. So that’s a great organic way to grow developer communities. And yeah, you can find me on LinkedIn, Twitter and you can also reach out to me at my email, email@example.com.
Matt: Beautiful. Well, Tanay, thank you so much for hanging out today. It’s been a real pleasure.
Tanay: Thanks for inviting me. This was a lot of fun.