January 29, 2016
Dev Tools Digest – Jan. 22 – 28
This Heavybit Developer Tools Digest includes news from DeepMind, Slack and CircleCI as well as updates from Citus, LaunchDarkly and Convox.
How many of you read a post I wrote called "My Billion Dollar Mistake?" Show of hands. OK, for those of you that haven't read it, ask somebody about it. Because I keep getting messages about it, and I don't know, I think I wrote something interesting. But the concept in there is that I used to throw Hiten bombs and that's why I made a billion dollar mistake. You should read it. But I'm not the only one that throws bombs, my friend Casey Winters does too.
Now, I'm going to ask you how many of you feel this way about product marketing. You can be honest. OK, fine. I'll raise my hand, all right? He tends to be very sensationalist sometimes, like me, and he's very thoughtful. You should definitely read his blog. You'll see links in a lot of this stuff that I have . If some of you have your computers open-- Most of you don't, which is awesome, so you get to listen to me. You'll see more context on all this stuff because I feel like there's so much out there, and I'll get more into that.
What I was wondering after I read that he's said that, is "What is this product marketing thing? What are we all involved in?" For context, I do a ton of it. I do a ton of it all the time in all of our products. I've rarely had someone full-time focused on product marketing at my companies. Usually it was a marketing function and we never really got to a place where we needed to hire that, because we introduced it into our product development processes. I'll talk a little bit more about that.
Just like you do, you're like "OK. What's product marketing? Let me Google it." So, I Googled it. Then I got an answer from Wikipedia, it's some process for promoting something. Then I got deeper, and Wikipedia has this whole thing about what it is and lots of content. Then it turns out HubSpot, obviously HubSpot is ranked for the term "Product marketing," like they're ranked for every other freakin' term out there. They had an answer . Then Intercom, they're not ranked for everything but they're definitely ranked for a lot of things "Product." So, they had an answer. Then they went deeper. Then my fine friends at Drift, they had an answer too.
They went deeper, and just like a Hiten bomb, I was like, "I don't know. I really don't know how to describe this. I know what I do, I know what we do on our teams, I know how I think about it. But I really don't know anything yet from reading all that stuff, because there's all kinds of different opinions about it. Except that if I were to summarize everything I read, product marketing is a lot of crap to do."
I needed to come up with something for all of you and be like, "OK. How do I think about product marketing?" I'm not a full-time product marketer, but it sometimes feels like I am. I love product. I probably love product too much, way too much. What is great product marketing? How do you know when you meet it somewhere, or how do you know when you do it? To me, great product marketing is simply customer obsession. The reason I say that is, back in the day you were in a world where the business needs you had or the product features you were creating were really upfront.
That was when you had a lot of on-premise software and a lot of businesses that were buying based on a checklist of features. I'm sure all of you know by now that a lot of that has changed, and I'm preaching to the choir, but to me it's all about customer needs and customer benefits. Discovering them, speaking of them, etc. So the best summary I can give for this is "Your customer is all that matters." I know I'm not saying anything revolutionary yet. Maybe I won't today, I don't know.
But what I am going to say is, to me, product marketing is all about customer obsession. That's what great product marketing is. We, at my companies, now have developed our own three steps to customer obsession. I made it up for this talk, but I really took in everything I know from the last 18 months to two years creating a new product, and coming up with this. Because I thought that all the crap out there was just a lot of to-dos and a lot of things that you had to do.
It felt like the customer obsession was missing, the part where you get obsessed with the customer to the point where you understand their needs and you also can map those to the business needs in a way that is focused on them. Because it is all about the customer.
First of all, work backward. Any of you heard of "Work backward?" Or "Working backwards?" OK, cool. Great. This is going to be fun. First what I'm going to do is break down these three, and then I'm going to give an example or two or three going through the framework. So, this is Amazon. They're really well known for the working backwards process.
They, I believe, invented it over there and it's all about working backwards from the customer. I'm a historian for this stuff, maybe I'm too geeky about it, but I really think about the products that they've had that have been successful and the ones that haven't. When I think about that, if you think about AWS as a product, they were able to predict the future.
This is the process they used to predict the future. If you think about their phone, I don't even know what it's called anymore, they were able to build a product where they were trying to track your face and all that stuff. What it felt like their thesis was is "We'll build different APIs so that people can build games on this thing because games are a big thing that attracts people to phones. Because developers get attracted to them and there's a reason to use this phone." If you think about the thesis, it was a reasonable thesis at the time. It just didn't make sense. Nobody wanted a fourth or fifth type of phone, or operating system.
It can fail, obviously, just like everything else. This is a detailed view on it. I think one of the key things is about how they iterate on a press release and they create that very early in their product development process. In fact, it's one of the first things they create, while most of us it's one of the last things we do. I've got some examples that are more modern a little bit later.
They have a whole breakdown of how they do the press release, and the key thing about this concept and the way they do working backwards is that they're taking inputs from customers, customer interviews, customer feedback and all that stuff, and creating this. In my opinion, product marketing needs to be heavily involved in this an/or own this document early on in the process. I'll talk more about what I consider more modern ways to do it, because quite frankly many of us are not actually chugging out press releases all the time.
The second step is what I would call "Engage, then launch." I am supposed to talk about post-launch, but what I really need to talk about is customer obsession and how you get to a place where you get to succeed post-launch and iterate. That's the whole thesis.
Has anyone heard this rant from me about early access versus beta, ever? OK, good. Fresh room. I get to inundate you with my concepts. This was a pet peeve of mine. The reason this one's a pet peeve of mine is because I see so much out there with this word, "Beta." It implies "Buggy, I shouldn't use it," and things like that. "Early access" implies that "It's special, and it's for me, and I'm getting access to it." I have long rants on this, I don't want to go off on it now. We made a nice little graphic about it because I wanted to make sure that the gospel is spread, especially this one.
It's super important to me and I think people are catching on, you're seeing a lot more early access. Google is doing a lot of stuff and they're calling it "Early access" now instead of "Beta." I'm not sure when early access gets played out, I am already working on the next term. I don't know what it is. Maybe it's like "Red rope line?" I don't know. Anyway, I'm working on it.
But this is really important because when you say it's not buggy, and it's early, people are more likely to share their opinion about what it is from a product standpoint not necessarily from a "Here are all the bugs that it has," more from an opportunity standpoint of "Here's what it could be." That's really important early on in the process, so that's why I like calling things "Early access," and what it turns out to be is that we start creating processes where we have a private early access, and then we have a public early access.
The public early access is very much a timeline product launch. Private early access is what you want to do to learn from customers while you want to keep things private and a little more quiet.
So, written a bunch of stuff about early access. I can go to town on the concept and why it's important, I have a bunch of blog posts out about it on my blog ProductHabits.com. But this is really what's important, when you do an early access you get to ask people questions. I'm a fan of open-ended questions asked the right way. I'm not saying you go ask a bunch of open-ended questions the wrong way, but this is one of the key questions I like to ask when I do an early access process.
The question really is-- This is an example for one of the products I built, FYI, where we wanted to figure out what the number one challenge with documents is that people have. So we asked them, "What's your single biggest challenge when it comes to creating and sharing documents?" Then we added this description to really hone in on getting people to give us as much information as possible. We had folks write essays about their problems in this type-form fill-in box.
This is one of the key questions we use for early access. You can do it for any product or business, B to C, B to B, hybrid, whatever. This is another one of my favorite ones because in the survey itself, or in the application for early access, you're basically asking them "How are you going to help us? How do you want to?" Then we had a tremendous amount of people, fill out everything except that one if they're not local. Because they want to help and we're asking them to help us build this feature or this product or whatever. Also, this process works for a feature or a new product, it doesn't matter.
This is one of my favorite ones. Many of you might have heard of it as "Fake door," a "Painted door," a "Smoke test." I recently talked to a friend that used to work at Netflix, and he's like "We called them 'Painted doors.'" I'm like, "OK. That's a lot more positive than some of the other concepts here." One of my favorite ways to engage and then launch is basically create these painted doors.
The idea here is basically explained by Optimizely where you just add a fake button, or a fake link. Sometimes you can A/B test it and then you can see which one people care about more. Whether it's the words, or even different placement and things like that. It's a very good tactic. I know many of you are like, "Yeah. I don't control that part of the product or anything like that." One of my pieces of advice is "Go get control over more of the areas in the product where you can do things like this, or make some friends on the core product team so that you can start testing things like this."
It's a really key way to do it. More information on this here, I think the biggest thing for me here is it helps you take care of a lot of the risk that's in the business earlier on, or risk with the feature, and that's all I want to do with all these processes. How to reduce the risk that, whatever my message is that I'm going to put out in the world, resonates and gets people to take the next step, which usually means sign up, use this feature, adopt it. Understand whether it's for them or not, which is also important. This testing really helps with that.
This is the last but not least step in the process, it's probably the most boring one and the one that I personally used to avoid, until recently. This is a template, that my Co-founder Marie, and I came up with in order to just do postmortems. This is a generic template on postmortems on anything. If you don't do postmortems or retrospectives or reviews that are reflective about what you did, you're not going to learn. Like period, full stop.
This is baked into everything we do. Sometimes I'll do a tweet or tweet storm, and I'll write up a little postmortem because it's something we want to repeat or learn from because something interesting happened. That's the way we think about it, and I highly recommend you take some concept like this and do something with it. As promised, I'm going to give some examples. None of you know about Product Hunt, right? Cool.
One of the things that I find really interesting when it comes to where product marketing has gotten to is the fact that you need to engage and there are a lot of communities out there. Product Hunt is just the example I'm going to give because it's interesting, and we have a deliberate process just like we do for many things.
This one is not three steps, I think there's five here, but basically it's: 1) We want to find out what works and what doesn't in the community 2) Discover how to engage with people 3) Make friends by contributing 4) Be timely and relevant and 5) Add value before we actually promote what we have. This is what most people do when they're marketing and trying to "Hack" Reddit, or whatever the words are, and stuff like that.
But now that it comes to communities you actually have to get really involved, so there's a number ways I've done that on Product Hunt myself. I created a collection that's called "Free stuff for startups," and it's the most popular collection on Product Hunt. I just did it on a whim because I know people love free stuff and Product Hunt has lots of free stuff on it. That's one way that I contributed and added value. I think it's one of the biggest things I've done on the platform because now every time I add a free product to it that's on Product Hunt, it gets emailed to all those 6-7,000 people that subscribe to my collection. There's those kinds of mechanics that you're looking to understand.
OK, so I'm going to give my examples of these three steps in action. We launched something actually more recently on a whim, and it was an integration with my product, FYI. W hat we do with FYI is we help you find your documents in three clicks or less across all the cloud tools you use. We launched an Evernote integration and it bombed, we usually do pretty well but it bombed. There's a bunch of reasons it bombed, and I'll get into that, but we just didn't do our deliberate process and we just wanted to launch it. We didn't have expectations or any goals for it, and that was all on purpose. Or at least, that's what I'm saying now.
We did another one that was for Product Habits. It's called Product News, and it did OK. It did OK because the product does not have retention and we knew that, but we wanted to launch it because we built it with someone called "A maker" and we had fun building it with him. We got a bunch of sign ups from an email list standpoint, and we grew our email list with it, but this was a very traditional launch it and know it's going to die. I still have to update the damn site every day, but whatever, because I just feel the need to. It's basically a site where you go there and you see product news, and its at ProductHabits.com/news. You'll probably see that I haven't updated today, I've been busy.
Then last but not least, this is when we actually launched something we were much more thoughtful about and went through a very deliberate process that I'm going to talk about, and it worked really well. It was basically a template gallery for product management that works across all the tools that you use. We have templates from everything from Docs all the way to Coda or Notion, if you're into those products. We have it very well organized and all that good stuff.
So, what we did is we worked backwards. That was the first step. The way we worked backwards is we basically created a Sketch template or a mockup of a Product Hunt page and we filled everything in. It's very similar to the press release process that Amazon would use, and we filled it in as if this product was live and these were the characteristics of the page. This is the copy, the images and everything.
You can see this version is a little different than the launch page here that I showed you, because we iterated it after we did this and after learning. The way we did that is we actually ran user tests from UserTesting.com on the flow. The flow was, people visit the Product Hunt page, they click on "Going to our website," as you can see here, and then they go to the website and we ask them a bunch of questions. We asked them a bunch of questions. These questions and about 10 videos of people walking through it and answering these questions, got us to iterate that fake Product Hunt page, and learn exactly how people think about the words and the images and the way we framed the whole product.
You can do this for anything, it doesn't have to be a product. You can do this if the way you're doing product marketing and marketing in general is off of GitHub, you can do this if you're doing it on Reddit. There's no excuses here on doing it.
Product Hunt is probably a good place to start just because I just gave you a template that the designers can take and move around, or you could too because it's in Sketch and Sketch is easy.
So, this is the way I look at it. If you care about customers, product and time, you'll work backward. Because it'll just save you a lot of headache from launching and then learning. I like to learn and then launch, because then it's likely that whatever your goals are, they're more likely to be met. If you have goals with some of this stuff when you do the user testing and things like that, you can really align with what you're looking to achieve.
Then we have "Engage then launch." This one's really interesting because we know this community really well, this specific launch that we did worked out really well, and here are the ways that we did it. One thing we did is we knew from the user test exactly what people cared about, and we knew how to think about the product marketing of it. A simple example is when you put a bunch of companies or a bunch of people in a tweet, they end up retweeting it or liking it at least, and then it gets more play on Twitter.
We knew that and this product happened to be a product where we had all these folks' templates in the product. So, this is what our tweet was. It had decent if not great engagement, it got us a bunch of clicks and everything. Then we did another one. This one's interesting because I tweeted something, and it's just a statement, and then I threaded it and the second tweet was basically the thing I really wanted people to go to and promote.
These are the strategies just with Twitter, we would play around with it and understand how to engage and go after it. These are just good examples of how you get to engage and then launch. We think about these tweets and things like that very early in the process so that we're baking it in just like we would the working backwards process.
Last but not least, in the Product Hunt community, people want to give you feedback. They want to sit there and tell you everything you're doing right or wrong. It's a very happy community in general, and we know that about the community. It's much different than Hacker News. Definitely got roasted on my billion dollar mistake on Hacker News, but that's OK. I'll live with it or cry in the corner. So what we did here is we basically made it very deliberate about asking for feedback, and really knowing and studying what comments get the most upvotes by the makers who post something on Product Hunt. We reverse engineered it and figured out, "OK. This is how we need to interact with this community."
This is near and dear to my heart. This is a lot more on the product side but it is relevant to product marketing, and it's basically this process. I like to use to engage in launch, which is "How do we engage with people even before we launch something?" I know a lot of the things I said involve early access and things like that, but there's just a formula here that I like to use. Start with step one.
Whatever you're building, however you're thinking about it, there's a first step to take. Make it a baby step and then basically from there learn and iterate really fast. I'll give a very tangible example. A buddy of mine, Bryan, who likes to watch some of the stuff we're working on and decided to finally sign up for my product after a year, basically tweeted this. This is FYI, this is the interface, it's actually a screenshot of him using the product and performing a search and finding a document. We're testing something. So, when you hover over each item you'll get this little plus button. You hover over an item, you see the plus button, you decide to click it and you think something's going to happen. We don't know what your expectation is, we can only imply that there's some expectation.
What we decided to do before we built the feature was take a baby step. This baby step is basically, show it on hover and make people click it, and then make them tell us what's up. What we like to do is once we know we're going to build a feature we start playing around with things like this, and then we make it a product marketing function because we need to understand what people think about this.
This could be a product function too, but we like making it a product marketing function because it helps us figure out how people think about this in their own words.
So we wanted to know, "When they hit the plus, what was their expectation?" Is it going to align with what we are going to do in the product for them? Another thing we did here is we actually tracked how many of these pluses were viewed, how many people hovered on the pluses, and how many people clicked on the pluses. That gave us a nice little funnel to understand whether we were going to get the adoption we wanted, because in our case a lot of times we think about product marketing as also an adoption thing. We should be focused on adoption. If we're not focused on adoption and the work we do, we might not know its impact.
Then last but not least, document what happened. Again, this is the template, and that's my face in it. This is it partially filled out. We'll take our template and then we'll fill it out a little bit just to figure out "What are we trying to understand about this initiative?" Then we filled it out all the way.
This is my Co-Founder Marie, who I have learned a lot about documentation and process and writing stuff-- she's a writer-- and postmortems and reviews, not just in my work life but in my personal life. If you folks want to tweet anything, if you could tweet at her and tell her to write about her thoughts around reviews and postmortems and retrospectives, maybe we'll get some amazing content from her about it. But basically, this is the line she says "Postmortems let you objectively look at all the facts of a situation, identify what went wrong, and what can be done better next time."
Again, just like product marketing, here's a simple example of what I would call product marketing for postmortems. Something I've had to learn the hard way from her. We have a newsletter at ProductHabits.com. It's free for now, probably forever. I mean it, if you have questions and you want help, just email me. I have help now. John, are you in the room? I got a minute or two, I'm going to give you a hard time. Where's John? That's John. He helps us with Product Habits now, so he might be helping at least tell me which questions I should be answering on here, but yeah we're really excited about product, product marketing, product development and share a ton of stuff that's relevant. Thank you for listening.