On this episode of The Unlearn Podcast, listen in to Jeff Barr, Vice President and Chief Evangelist for Amazon Web Services (AWS), where he has been working for over 20 years. He is known for his influential blog, the AWS News Blog, which targets builders, developers, architects, CTOs, and students. Over time, Jeff has built a team of bloggers and implemented a streamlined content creation process. His approach involves simplifying complex topics to make them accessible to a wide audience.
00:00 - Introduction
02:55 - Jargon in the tech industry, a hindrance in distilling complexity into simplicity
06:40 - Bringing authenticity into your content
08:25 - The story behind the creation of the AWS News blog.
11:16 - Objective behind starting AWS News blog.
14:17 - Evolution of AWS News blog over the years.
16:57 - Scaling content
22:25 - Team size required to support content evangelism
24:54 - Building high quality content
30:40 - Moving beyond bits and bytes, feeds and speeds.
34:08 - Visibility into the end-to-end reader journey
37:24 - Tracking content performance through metrics
43:16 - Spaces to go for finding and building communities
46:06 - Thoughts on AI in content
“We need to move beyond bits and bytes, feeds and speeds, because it can get really boring and really dry really fast. Content has to be about people and how it affects them and what they can do with it.”
00:17
Hey, everybody! We're back with another episode of Unlearned AI. I still haven't been able to convince Kelly to kick these podcasts off, so I'm still working on that. Hopefully one day we'll get there.
00:38
That's because you have the radio voice, you have to kick it off.
00:42
This is gonna be a fun one because of the guests that we have. And Jeff, welcome to the podcast. Do you wanna just give us a little bit of insight into who you are and a little bit of how you got to where you are, and then we'll kick the podcast off?
00:58
Sure thing. So, firstly, I'm happy to be here and I always love sharing stories and a little bit of my career history. So, my name is Jeff Barr and I am VP and Chief Evangelist for Amazon Web Services.
I've been working for Amazon for a little over 20 years and I started my career deeply in tech, but very quickly found out that what I was good at was first diving deep into some kind of tech, some kind of complex system, and understanding it.
And then, over the years, the balance of how much of my tech side versus my communication and marketing side has shifted more toward communication and marketing. I still love and enjoy tech, if I don't get to build it, I get to understand and appreciate it.
01:44
Super, So, Jeff, what's top of mind for you? We always start with this question because Kelly and I have, like, tons of questions, but we always start with, like, what's top of mind for you? And then we kind of break into the podcast.
01:56
So, top of mind for me is always simplicity. It's so easy to write complex, hard-to-understand explanations. And I think some content creators almost relish the idea of, 'Let me make something so elaborate and complex that no one can fully understand it. And you almost have to be a wizard just to get it.' I'm always striving for what I want to be.
I want to explain things in a clean, clear, understandable way, without making it seem as if I'm talking down to the audience. There is a delicate balance; we're using just the right level of sophistication to get the point across without using too many words or words that they have to look up.
I have a global audience for the content I create and am very aware that I have an audience where English is not necessarily the first language. So, I always try to think of my audience and my customers when I'm creating content.
02:55
Yeah, I think that's a great point: jargon, in particular, in the tech industry. But also, when you look at, like, a kind of consulting analyst space, where you can get a lot of good ideas, but it's also often laden with jargon.
And so, people have to go Google search, and they, or they just quite frankly, don't bother to understand what is being said, right? Because it takes a lot of work to translate.
That's hard, right? Like distilling complexity into simplicity is a talent, right? It's very hard to do. So I think why a lot of content doesn't do it is that it takes more work, right? Like if you want to write a tagline for something, those three or four words may take months, versus writing an article that has 800 words to explain what you do, which may be an easier task.
03:50
Exactly. It's those short, very fundamental pieces of content that are so important and so difficult when I'm writing a blog post because a lot of the earliest content I wrote before computers and word processing was a thing was often actually on a typewriter; I wanted to get things right, and I tend to write linearly.
And I find that getting a title correct for a blog post is one of the most difficult things. I think, if you could look at any of my posts, and if you could see my character-by-character, word-by-word edits, you'd probably see that a lot of the posts have three or four titles that I will very quickly iterate on.
Run them, buy my audience in my head, and I will not proceed past the title until I'm happy with it. And, as far as I'm concerned, it's final at that point. So, it is that simple. And I do think that people who don't design or create as part of their job or their hobbies often think that you start with nothing. And then the first step up from nothing is the simple thing.
But that's usually not the case. You usually start from nothing. You create this very elaborate solution to something and then you need to whittle that back down to what are the essentials?
Sometimes I think you don't fully understand the problem you're solving until you've built that elaborate solution. And then you get that bit of insight and say, OK, now that I see I've made the hard, complicated one, I can make the simple, much more easy-to-understand one.
05:20
This is going to be amazing, an amazing podcast because, like, I've been so this is what I've learned, right? I went on to create the Sunny Sideup podcast, which was a project to start the demand gen efforts for this company called Demand Matrix.
But selfishly, I just wanted to bring all the people onto the podcast that were amazing creatives and that were amazing content, that were like marketing leaders, right? And that's on that podcast.
I learned about, like, Bill Bernbach and all the amazing, like, thought leaders (or leaders, I would call them) that brought the creator movement, as we call it, to life. And this creative movement, I feel, has been kind of compounding over the years. And it's kind of got to the point.
The point where, like a lot of executives used to, outsource their LinkedIn posts, blog posts, or just their external communications. And they've started to, like, do it themselves and you can tell--not kind of--you can tell, like, who is doing it themselves and who isn't. Because, like this, when they speak from within, they, like, it looks like they've launched a mini product in the market.
06:40
I have to agree; authenticity is so important, and being yourself, for you know, bringing your personality, your background, your experiences, maybe bringing in other aspects of your life that are not just about your job and saying, 'Hey, I'm a whole person and I have other things that I do and that I'm interested in that affect me besides just my job.' I think it helps the audience to relate to you. And, you know, earlier in my career, I think all the professionals looked so...
Slick and so polished, and you think, 'OK, everything is going exactly right for all those people, and nothing's a challenge, and they're all perfect.' Then, you get into one of those roles, and everybody then treats you like that. But the reality is, we're all just learning and growing, and hopefully adapting and changing as well. I have this personal thing about never getting stuck in a rut and never getting fixated on any particular technology or iteration of technology.
And that's helped me to stay fresh and hopefully young versus I met a lot of people early in my career that would just get stuck on one thing and say, 'This is the thing that's gonna keep me going for the next decade or two.' I don't subscribe to that at all. I'd much rather be on the leading edge and always trying something new. Maybe it's not fully formed, maybe it's not quite there, but I'd rather be there than just stuck a generation or two behind.
08:06
It's great!
08:08
And who is your audience? Because I feel like you probably have a fairly large audience; you said it was international. But what about the persona? Like, are we having business people read it? Is it more just technical people? To whom are you reaching?
08:25
Sure, so even before we get to that point, I should explain that the main piece of content I created is called the AWS News Blog, which I started writing in late 2004.
So, we're coming up on the 19-year mark for that. I launched this blog when we were first getting started at Amazon with our infrastructure web services. At that point, I was the only person doing any kind of marketing activities and developer outreach. And I wanted to reach a wider audience.
I always really thought of my audience as simply we're builders, and a builder might be a hands-on developer. It could be an architect who's maybe creating plans and diagrams and abstractions for others to use. It might be CTOs that are trying to understand and evaluate a technology; those are the folks I think of as being my audience. But this starts at the student level.
It's quite fascinating sometimes where AI will be. I'll be on campus, meeting with some students, and they'll tell me how they were just learning about the cloud and learning about AWS. They did some searches and they happened upon my content and found it useful. It's very gratifying to be able to create this body of content.
The years, and of all the rewards you get, hearing that it helped somebody on their journey - that it helped them to learn something new, it helped them to accomplish something. It helped them to grow in their career, maybe make their life a little bit better for them and for their family - that's what I find just really gratifying about what I do.
10:01
That's amazing. And, so, did you say you started this in 2004?
10:07
That's right; it was late 2002, November 2004, when I put my first two posts online, basically saying, Hey, here we are! For the first 10 years, I was the sole author and we didn't have a team around me.
As our launch schedule accelerated and we built the audience, we slowly put a team together. In those early days, we didn't even have an editorial calendar; we didn't have a process; it was simply my colleague was just emailing me or catching me in the hallway and saying, 'Hey, Jeff, we're about to launch this cool thing, and it would be great for you to write a blog post.' So, I managed everything pretty much manually through my inbox, and that worked until it didn't.
Then, we started by hiring a team and having an actual systematized process where the teams get to create tickets, we now have multiple bloggers with whom we can all discuss and agree on who's going to write. We have a draft process and a review process, but for the first 10 years or so, it was just me and my inbox, and my blogging tools.
11:16
Wow, that's Amazing. And what were your goals with the blog? both now and back in the day when it was just you—in terms of, were you trying to drive awareness? Were you trying to attract new developers to work on AWS? How did you conceive of it when you started it?
11:35
So, that's a great question. And before I launched the blog, I had been doing an awful lot of travel my family was very young at the time and they were very understanding of the fact that I was traveling a lot I was thinking about how do I scale my effort and be beyond simply how many events can I accept invitations to and how many places can I be in the course of a trip?
So, the thinking was, we're doing some amazing things at AWS. How do I get this out to developers all over the world? And I thought of it in terms of my colleagues building all these cool services. I think they're cool, they think they're cool, they were designed to meet the needs of the audience.
And I want my audience to simply understand and appreciate what these things are, and then hopefully say, 'That sounds amazing! It solves a problem I have; I'm now empowered to do something with this.'
12:32
No, I'm still stuck on that. You've been doing this for 20 years; you know, I don't think I've met anyone who's been blogging consistently for 20 years and been able to build this thing. OK, this is great.
12:47
You know, one of the lessons for me is that it can take a long time in your life and in the arc of your career to find the thing that you are good at, maybe even the best in the world. And I'm not saying I'm the best in the world at that, but I've found my niche and that niche. I was 40-something when I launched this blog, and usually, people have kind of found their role by then.
But I'm not saying that I didn't have any actual, like gigantic crises or failures, but I was doing things that, in retrospect, were all leading up to being able to do this. But I'm better at doing what I'm doing now than I was at doing any of those other things.
Keep yourself a little bit open, a little bit flexible, and then sometimes you're doing one thing and this opportunity presents itself that's a bit of a fork in the road. And I tend to almost always say the universe has engineered you to get to the point where that fork is the right thing for you to do.
And sometimes others can maybe see something in you, you know, your skills, your attitude, whatever you've got, and say, 'I know you're doing this thing right now, but have you ever considered this other thing?' And quite often that's because others can see things in you that maybe you're so close to that you can't.
Stuck in your view of yourself, you can't quite see, so almost always take that fork in the road.
14:17
That's great life advice. Amazing!
So, how has it evolved in terms of now that you have, like, a team? Right? Has it become more structured and sort of part of a larger, orchestrated strategy across developer relations?
14:38
It has a little bit, but we still try to be informal, in the sense that we don't get too corporate and we don't have an approved list of vocabulary. And what I have is a little alarm kind of in the back of my head, or almost like this little mini Jeff watching over what I write.
And when we get to that jargon point, when it's just a bunch of buzzwords strung together, there are times when I'll start with some of our internal descriptions, and it's sometimes a little bit too buzzword-laden. You type it, and then little mini Jeff says, 'You know what? That's just a bunch of BS; that's not helping the customers. If you can't satisfy yourself that you're communicating at a high quality, you're never gonna satisfy anybody else.
So, you need to be a critic of yourself, but you also need to be understanding and say, Well, you're doing something pretty good here.
15:36
Just, just for everybody's info, Jeff has almost 75,000 followers on LinkedIn, just so that people understand the context of what we're talking about here. Jeff, so, I feel like what you've been able to do is go from content creator to content operator. Would that be a good way to explain your journey?
16:01
I'm not quite sure what the difference is. I may be the creator, you know, we make that initial content, but the operator says, 'I think, how do you get continued value out of it?' And so, I do think that being adept with social media is part of being successful here and...
That's something I still think I'm learning to be awesome at because there are so many different platforms and you can spread yourself way too thin and try to have a footprint on every last one of them. But being everywhere is hard because you need to understand your audience, which is different from platform to platform. You need to be able to not just...
Be there, and you can't do hit-and-run posting, right? You can't just post something there and not pay attention to replies, comments, or feedback, and then...
16:51
There's no engagement type.
16:53
Personality, and unless you understand that, you can't be on all of that.
16:57
What I mean, by the content operator is like, like you were, it was just you. And then you said, Well, how do I scan myself? But how, as I scan myself, how do I scale content for my company as well? And so, I'd love to have to just unpack a little bit of, like, Well, how is your team organized and, like, how do they think about content every day? And, like, what type of support system do you have around you? Because, you know, there's a lot of, let's call it, content resources out there, you know. I would say, and, and, and they listen to all the people that are also trying to figure this out, and some people have actual systems for publishing and producing, and there's an end.
In their mind, where they're like, 'Do this, do this, and then here's a destination, right?' Not all content today that's available on social media has some sort of destination. A lot of it is somewhat nowhere, but a lot of it is debated. So, I need help understanding the resourcing that you have behind you to be able to direct content and engage your audience because, like, an interview has, I don't know, how many thousands of products now that you can offer.
18:08
We're at the multiple hundreds of services; I think the latest number is 329 services. But to that, you have to add the fact that we launch these individual services and then we do thousands of additional feature additions to each, and then people...
18:22
Build on top of that, and like they, this circle goes on over and over again, right? So, you're like, I love how this podcast started because your job, if I'm wrong, correct me, is to simplify so people understand, and then they can take their current problem, match it to the service, find a solution, and get going to the next problem.
That's
18:46
Right, and so one of the fun challenges we have is a set of leadership principles of Amazon that are available publicly; people can search and find them. And one of the things I did as we were scaling up from just me to a whole team of bloggers was to ensure that, right now, we have, I think, eight bloggers located all around the world. Some, like me, started with writing in English, while others have English as their second or third language, but we all work together.
And we all bring our unique backgrounds to the team. We have one blog manager and she's in charge of helping our internal constituency. All the teams that are launching services work with her to create a ticket for the new launch. They say, 'Here's the launch, it's upcoming, here's the launch date, here is the internal documentation, and here's a very important Amazon document called the PR FAQ, which is shown after the press release and press app question. Very.
19:48
Familiar with that,
19:49
Right? So, we have access to the service, and these teams are continuously creating tickets. We don't write a post for every ticket; we have some discretion within the team to decide what we think the audience needs at any given point. And, as we're writing the posts, we are always actually thinking of the customer. We always get to use the service. Hands-on AI; I always tell the teams, 'I love you. You're all awesome. I don't believe a word you say, so I have to use all these services first.' Cool. Got it.
You have to have authenticity in saying you can't say this is how they told me it works. It has to say, 'I used it and this is what I witnessed and this is what I saw.' And so, we always go for that 'hands-on' approach. So, during the writing process, what I did is, as we were scaling up, I took all of the Amazon leadership principles and kept all of the titles intact, then I rewrote the definition of each one to be very specific to how we think of it as a blog.
Blog. My favorite one is frugality, which is our second leadership principle. And in a lot of situations, you think of frugality as money or resources. But within the blog team, we think of frugality as being very respectful of the time of our audience. So, with a very large reader base, every extra word we put in multiplies out to a lot of wasted time.
So I always tell our bloggers, We want to be concise but detailed. We're going to choose our words with care and we don't want to overwhelm the audience with too much content. It's very easy to say, 'I must have at least 1000 words or 2000 words and try to strive for as much as possible.' We try to keep things tight when we're getting ready for a big event. Like every year, we run an event in Vegas called Reinvent.
As we're planning, we'll set ourselves up with a word budget and say, regardless of how awesome and rich this particular service is, we're gonna give it, let's say 750 words. And within that, we don't strictly hold ourselves to it, but we use that to set expectations with the various teams that are building because you could easily take any of these services and write thousands of words and you could write a part one, part two, part three. We say, yeah, we know how to do that, but that's not how we choose to work. We're gonna respect our audience's time. We're gonna keep these posts to 750 words and maybe five screenshots, so we'd rather have them finish the post in one setting, versus leaving it on a browser tab where they'll probably never get to it again.
22:35
That's amazing. And, if this question is due, let me know. But, how many people are there that support all this content evangelism? That's kind of what I put in that bucket.
22:47
So, we generally don't give out different internal team sizes, but you could look at the blog and count the number of different bloggers. Right now there are eight of us. We tend to have me full-time charged with content creation, while my other colleagues have other duties.
In addition to writing a blog, that's what I was.
23:06
Kind of trying to get to it because it sounds like you're the product manager of the content, right? And then, and then there's a process that's running, and then you have contributors. Do you also engage with external contributors? Like, people that are not Amazon employees?
23:22
We currently do not, and there I think there are a lot of great other content creators out there. But one of the things you think about when you're in a popular situation is, How do you scale the effort? And there are a lot of things that I could do that I realize would be an open floodgate situation, where the first time we had a blog post from an external party we would accept it onto the news blog.
We would get hundreds of other requests that say, that looks great. I would like to be there as well.' And maybe it sounds a little bit weird, but having to tell people 'no' too often is painful. I like to see people succeed, and I like to be able to say 'yes'; having to say 'no, I'm sorry, we can't after you do that enough times, it's like this is not enjoyable. And so, we don't even take submissions from inside the team unless you're part of it.
Inside the company, unless you're officially part of the news blog team, and this goes to the point of, against scale. And also that authenticity; sometimes we'll get very busy and we'll get that back to the team and they'll say, 'OK, and with the best of intentions, they'll say, 'OK, Jeff, you don't have to write it yourself, but we'll ghostwrite it for you and you can put your name on it.' We just don't do that if we fully believe in giving credit where credit is due. And we're never gonna allow ghostwriting because it's not authentic.
And if someone else wrote the content, I would want them to get the credit for having done so. Yeah.
24:54
Went through a phase personally where I got caught into, like, Let me post every single day, every single hour, right? And then I realized this is all crap; even I wouldn't read it. And so now what I've done is, because of my interest--and this is what a great I would say--this is something like a podcast, master classic content in the debate, right? Like, my interest is to work with other executives that are in partnerships and help elevate them, right? And I've just found that the best time to engage people like that is either after four PM, right? Or on Saturday mornings and Sunday mornings. So I tested all of this content.
At times, I realize that if I only do one extremely high-quality post, which is very natural to the current conversations that are happening in the market, it allows me to engage, grasp, and get a feel for, Where are people at, and what do I need to do to build them? So, my process is, I'll have, let's say, 50 conversations.
In a week, right? Or let's say 45 days, like, right? and...
26:04
And Asher doesn't work; he just exactly.
26:07
I like to wait.' I'm over here 'cause I'm like, 'OK, people will be like, 'Well, this guy just talks all the time, right?' But there are some days where I'll have like 10 meetings a day, especially as we get closer to ASAP. I'll take the most interesting conversation that I had and really think about, 'Where do I stand on it based on my values?' and just post about that.
And what I found is that, over time, the audience that I care about and I'm passionate about helping also starts tuning into the time and they look forward to it. Then I engage with them and we work some things across. Then I take what I've learned and go back to my team to share with them so we can build features, products, or whatever.
What else do we want to do about that? But this is limited, but very high-quality, content. AI is AI; it's the right way to go, and kudos to Kelly because she and I had a debate about the content quality last year, and, you know, just like most things, Kelly won. So, I said, 'OK, well, let's try this out.' And it served me well.
27:18
I like that idea as well. And I, I remember in days past, my colleagues would come to me and they would always say, 'Where, where's your content calendar?' And you make sure that you always post on Mondays, Wednesdays, and Fridays. And do you have time? And I don't. If there's, if I have nothing to say, then I'm just trying to say the world isn't gonna miss me. There, a day might go by and I won't say anything. And when I do have something to say, I think there's a little bit more respect for the fact that, well...
You didn't just try to fill up the pipeline and keep yourself, keep your output full. Just when there's nothing to say, well, just be quiet and let others have their turn. I have nothing wrong with that. And I'd much rather post a little bit less than necessary than too much and have people start to tune you out and say, 'Well, it's kind of obvious when people do fillers, best-ofs, repeats, and low-quality content.'
Not thoughtful content. Right. I always like to impress myself with whatever I do and say; would I find this useful? Is this helpful? Maybe it's a bit entertaining, but if it's just OK, then for one good example, for a while I used to write April Fools' posts and I had a lot of fun putting those together and I actually would...
Writing them, and as I was putting them together, I was enjoying the level of humor I was putting into them at a certain point. I couldn't impress myself anymore with what I was putting together for those posts, so I just stopped doing them. I had to be my quality control and say, 'Well, if I don't like it, then no one else will either.'
And I, I think you have to apply that to pretty much everything you do. Do.
29:03
Do you hang out with others like you at your level in the Seattle area? I want to take a name; it's somewhat at your competitor's company, but I just wanted to see if it naturally comes up and if it's OK to share.
29:20
Sure, I don't get to hang out with a whole lot of folks. I mostly work from home and connect online with people, but I sure will let you know what you're thinking.
29:29
So, I'm thinking about Steve Clayton. Do you know Steve Clayton by any chance?
29:33
I don't, but if he's here, I'd love to meet him sometime. Super!
29:38
Is he the chief storyteller of Microsoft? And it's a story very similar to yours, right? Almost 10 or 15 years ago is when I met him. That was the first time I learned how to put a story together. At that time, Microsoft was very product-focused and there were no stories, but the work Steve did was to go and sit with engineering and product teams and unlock what the story is. I think you would enjoy meeting him.
30:14
OK, that's interesting, Asher. It sounds the way you're describing it. His role, though, sounds more like branding, which obviously would relate to what Jack does, but it's different. Is that correct? Is he more, is his role more to come up with these core brand stories?
30:30
Well, maybe that's the right term, but you know, his work always brings humanity into what's happening, and it's like...
30:40
I think the important part here is that there are people inside of all of this, and if it's only bits and bites and speeds and feeds, it can get really boring and really dry fast. It has to be about people and how it affects them and what they can do with it. And then it has to be beyond the press release quotes. If you look at all of my posts, you'll see that I rarely use those standard quotes in the press release. The new Amazon blah, blah, blah is amazing and we sure hope that we'll use it some time, and then it's like, you know, says so and so from the big important company; I never use those because there's...
I don't think they had any value for the readers, and they're so flat and so predictable. We just have almost a policy.
31:32
They do. I'd much rather have an in-depth post if someone at one of those companies uses a service and figures out a cool way to do it, or they build something awesome with it. I'd be happy to link to something like that. You know, if Company XYZ looked at this new feature and here's what they've built and here's what they have learned from doing it, that's far more interesting than just one of these bland endorsements.
32:04
I remember doing a podcast with somebody again, very similar to you who was an SFV AI. Always feel like you guys all need to do a retreat together. You know I just think about that, and you guys can come up with your podcast and do your summer together. You know, I think that would be pretty cool. But I'll make the intros after this call. I think you would enjoy being because they were like all, every one of you was all about like, 'Hey, how do I take this? Tell it very simply, specifically, beautifully, like the whole thing is packaged together so that it's amazing.
Experience for the reader. And yes, you know, some people don't like pictures and some people like pictures, but like you all three of you are very conscious of, like, how do I do this? But how do I do it at scale? So that multiple audiences, living in multiple geographies, can all consume it. Now, is your blog translated? I'm assuming it is translated into different languages. I'm curious how many languages you have translated it into today.
33:00
So, we do some internal translation and some of it's done, I think, by semi-automated means with the review, and others are done manually. I just came back from Istanbul, Turkey, a few days ago, and one of my colleagues there, one of the many things she does is she takes the English language posts on the news blog and then translates those into Turkish for us. And I didn't know that that was one of her duties, but it was great too.
To hear that now was great and good to hear that we have a good audience in Turkish. I find it interesting though, when you talk to developers around the world, you always want to meet people where they're at. And to me, that localized content is part of doing that, making that effort to give them what they need in the language they're most comfortable with. On the other hand, it seems like English is almost the universal global language of developers and sometimes I'll hear them say, 'Well, I read a lot in English to help me develop my English reading and speaking skills.' So, we give them both options.
34:08
So, Jess, speaking of the scale component, what visibility do you have in terms of the end-to-end reader journey, right? So, I think, with developer relations in general, but also with content, it's often hard to see what happens and what the impact is of the content. So, we can track views of our content and we know that people are reading it, and obviously, that's a good sign. But, ultimately, content has a particular purpose, right? So, it's another thing to say, hey...
People take this content and use it. And, of course, if your call to action is embedded in the content, it may be easier to track. But, I think in your case, and usually with developer content, it's not so easy necessarily to track the journey from a technology perspective. But I'd be curious how you guys look at that and how you try to see what happens next from the reader: where do they go now that they've read this content?
35:07
Wow. So, this is, I think, one of the trickiest topics. We certainly watch metrics internally and we know what posts and what kind of posts do best. And we try to understand a little bit about that; posts for certain kinds of launches tend to do better than for others. We do track all the standard things, the balance rates. And we do try to include a call to action on the page, but we don't get overly scientific about it because I. Developers see through any kind of overt marketing very quickly. And as soon as they think they are being marketed to in upper case, they say, my gosh, they are trying to trick me into doing something.' So you have to be very respectful about that. And yes, there will be a call to action at the bottom of the post most likely, and we certainly hope that they do that.
I tend not to look at any metrics whatsoever, which makes me an absolute pariah among my Amazon colleagues, where everything we do is to be informed by data. And the reason I don't look at metrics is I'm not sure what I could do differently if one post gets me 100,000 views and another one gets 200,000.
That's awesome. But there's an element of chance to how these posts get picked up online, and it's not always directly related to what's in the post or what we do. Trying to reverse engineer that is kind of maddening, and to think, 'Well, I need to write this by all the right rules to make it a bestseller.'
It's hard to know you, so sometimes the posts you write for a brand-new, really awesome, super-cool service do just fine. But then you write another post, a kind of medium-sized feature, which might be what I think of as a puzzle-piece feature, where we have a couple of different aspects of the product line and this new feature brings them all together in some way.
Those can be overwhelmingly popular and you don't even know that they're going to be until the audience looks and says, 'Wow, that, that, that now solves the final; that's the final piece that now lets me do a really good solution.' You don't always know that ahead of time. So you have to leave yourself some flexibility to see what's going to work and what's not going to work and not be too much of a slave to the metrics.
37:24
Did your top post do over a million views or impressions?
37:33
I don't share the numbers externally, so I don't think I can say that, but we do well. We do well. Now, the funny thing is, when you publish web services online, right? They all have a lifetime. You launch them, they grow for a while. most of the time they're gonna last in the indefinite future, but every so often when you...
Either deprecate an entire service or sometimes just some feature of a service. When we do either of those, we do it with a lot of planning, a very long time horizon, and a lot of care to help customers migrate to what would be the better solution. So, I've seen several times that the posts I write say, 'Dear reader, we know you've enjoyed this feature; sorry to tell you, it's going to be going away at some point. Here's how to figure out if you're using it, and here's how to move forward: those posts that are effectively negative tend to do well. So, the ones that take away from the audience sometimes do just as well as the ones that come from.
38:37
I do. So I have met, I think, of all the people because, again, I became a very big nerd about this whole content journey. After all, I just figured out, like, this is so cool. Look, if you can explain things so simply and, then I know the person and, and I always keep a tab off, like, well, who's got, like, who's done a post that went to do X hundreds of thousands of? So then I know one person and the, the, the number one post that they did got like 800K impressions. I was like, Wow, that is amazing, because, and I shared that on the podcast, is because people get reviews.
That's the reach of LinkedIn, you know, and so, like, it can be.
39:17
Very, yeah, but let's just pause there and acknowledge very clearly that the number of impressions or views is not usually the best metric to talk about impact, right? So there's probably usually a LinkedIn post if you post some personal story or crisis; it's probably going viral, especially if it's an attractive woman in the post. But that doesn't mean it's gonna have a lasting impact. Jeff could post something cool and innovative and it strikes up support people.
And then they go do it right. They go use the service; it, like, transforms their product. That's why I was talking about impact. I think views are one sort of indicator of impact, but it's not, and I think chasing after views is...
40:12
Not ag[reeing] with people to [just] shit after reviews. What I'm saying is, like, understand the magnitude of what you're dealing with because I don't want this again. You have all these entrepreneurs, right? They're like, yeah, I'm gonna go do this, blah, blah like, you know, and then they're like, I got 50 likes, but fantastic, right?' It's like, 'Well, there's a much larger audience that you need to dive into and understand the needs of, and like, it takes time and you still have to get there. but for recordkeeping...
Like OK. Well, who's got this? What does that feel like and mean, and how broad can it get there? Right. So, that's the only reason I learned this, but I...
40:46
I like to rely on it, and maybe this sounds a little bit weird, but I like more anecdotes and stories. So, when we have any kind of event that I go to, I'm much more excited by just hearing from people that say, 'Hi Jeff. You know, I read your post about this and it helped me and I used it to further my career and my life.' I just like those little bits of recognition that say, 'I made a difference in somebody's life.' And the interesting thing about living here in Seattle is that we have a pretty big tech population, but the city is kind of small. And so, if my wife and I go, like, walk down from the hill, go down to have dinner and we're out in the city...
Quite often, someone will just come up and say, 'Hey, I recognize you from your blogs and I love what you've done. This is cool, and it's just a touching human moment to hear from an individual that one individual says, 'You know, I listened to what you wrote and I applied what you told me, and I've now stepped up in my life. I've got a certification. I've got a new job.'
That's far more interesting and heartwarming than 10,000 page views or something because it meant that I changed somebody's life in some little way. I told them.
42:06
Comparing that to the last 20 years that you've been doing this is fantastic. It's just amazing to see that you found your colleague, stuck to it, and did it for a very long time and you're still enjoying it. You're talking about this stuff like you just started and we can feel the enthusiasm around it coming through clearly.
42:27
Go ahead. Yeah, liking what you do is so important. Like, enjoying your job and getting up fresh every day and saying, OK, what, what, what am I gonna do with this? And, like, what cool thing is in my inbox that I get to write about?' And, I have to acknowledge here that part of what makes this fun is that I have all these amazing creative ideas.
Colleagues are building awesome things to meet the needs of our customers. So, like, all this amazing stuff lands up in my inbox, like, 'Hey, we have this and this and this, and it's this cornucopia of things to choose from and it's this almost embarrassment of riches of, I can do this or this or this.' And the challenge sometimes is that there are so many great opportunities to share things that you have to just pick one out of ten and say,' OK, this is the one I will do today.'
43:15
Yeah, but it's Great.
43:16
Now, I'd love to circle back to what you were talking about earlier in terms of sort of the spaces out there that you go into to sort of build the community. Because I think, you know, for developers and developer marketers in particular, it's always a question of where do they go? And I assume Asher was referring to your LinkedIn followers when he said 75,000.
I wouldn't have thought that would be the leading forum for developers; I would have thought something like Discord or Reddit, but I'd love to hear your take as someone who's immersed in the world. Where are the communities that that sort of person is most hanging out in, and which ones do you find the most productive to go into and kind of share what you're writing?
43:59
So, I've found LinkedIn is the best place for me to disseminate what I do, whether it's a self-contained post where I just share, Hey, this is what I'm thinking. This is what I'm up to, or this is a piece of content that I'd like to refer you to, I'm getting far better results there than I get on Twitter, which I've stepped back a bit from. I do have a presence on Reddit and I'm happy to connect with folks there.
Trying a couple of different other things. I know my colleagues are dealing with, or are putting content on Instagram and TikTok, and they're having some good effects there with content that is specific to those platforms. And this is why I think you can only truly be at home on a few platforms, 'cause you need to understand the audience and kind of the dynamic of how you create and interact with the audience on each of those, to be legit. Like simply copying and pasting the same thing across eight different.
Platforms? Yeah, like, and some tools will do that for you. I wouldn't go anywhere near any of those.
45:04
Yeah, that doesn't work well. So, who, so what on LinkedIn? Do you find a lot of developers engaging in the content? Is it also like CTOs and tech partnerships people, or is it working developers? Is there a critical mass there that engages with your content?
45:22
A pretty healthy mix and one of the things I do like about that platform is that I do get information about the demographics, so I can see the mix of titles, I can see, you know, geographically where they're coming from and I find it helpful, but I would never look at the metrics and say, 'Well, I'm getting more CTOs and developers. Let me change my content in some way.' I think you have to be careful about getting into too tight of a feedback loop and saying, 'Well, I will not optimize this and somehow I will uplevel it because I think that CTOs will just want overviews and not details and maybe there's a good route to success there, but that's just not the way I would happen to do it.
46:02
What do you think about this whole content? I wanted to wait till the podcast got out of the water.
46:09
So, we didn't spend another 45 minutes on AI?
46:14
It's fascinating to see how quickly it went from something that experts knew about to something that everyone can use and benefit from. You know, a lot of these things have been in development for decades, but it was only recently that they became available. We kept thinking, Well, there'll be some way, someday when this goes from a small scale, great for limited projects and experiments, to suddenly everyone can use it and benefit from it.
It's fascinating how quickly that happened—earlier this year, like we were having dinner with some friends, neither of whom are in tech but are, technically, you know, capable. One person is a program manager, and the other person is a high school math teacher. They're the ones that came to me and said, Hey, Jeff, are you using AI? We're both using it in our jobs. It's like, Wow, this has not gotten out into the mainstream—regular folks are not just knowing what it is, but figuring out how to write prompts and get value from it. So...
Thrilling, and we should have to think of the implications in the longer term of what's real and what's been created. I do worry about the audience's ability to distinguish the two, and the potential to fool people is very high, and that scares me a lot. There have been some of these deep fakes; you know, you can deep fake audio and you get this panic call from someone claiming to be a family member: 'Hey, I've been kidnapped and you need to do this,' and the whole thing has been deep faked. That's very scary. And even within our family, we're thinking of little things, like a kind of personal authentication, not encryption, but authentication, between family members, to think: how can we authenticate that the person on the other end of the line is a family member?
47:59
I, so, what I was thinking was, you know how Twitter has the blue check mark? Now they're gonna have a green check mark if you wrote it; you know, somebody's trying the green check mark.
48:09
But, what if you use AI to give you a nice opening structure and you say, OK, this is mostly correct, and then you are effectively the editor for what the AI has produced?
48:21
The writer solves it, and then you take it from there.
48:27
Ah, I've done some experimentation, Kelly,
48:30
I've done some experimentation. I did one really fun thing where I said, 'Imagine you're a waiter in a restaurant; explain cloud computing in terms of menu items in your restaurant.' I got some fascinating, creative results out of that.
48:46
That does sound interesting. Although, I remember I used to write about an API and they always had the restaurant metaphor; it was like, 'go to the waiter, the API', and you know, I love that example. but it sounds like you are getting a bunch. I mean, I think in your role too, right? It sounds like you're getting an influx of inbound requests for content. Have you ever considered trying to have AI read all the inbound requests and give their opinion on which one should be pushed forward?
49:15
Wow, that's a great idea. But what if those requests are also coming from an AI, and suddenly we have this AI-AI communication?
49:23
Yeah, well, yeah, and I was talking about that. In another episode, it is like everyone on social media starts using an AI to talk for them, and then pretty soon it's just the AIs talking to each other and nobody's real. That's a real danger on social media.
49:38
But look, if you use Gmail, for example, but you're using Gmail and it starts doing these little completions that are probably, and hopefully, trained on your content - where's the boundary there? Right? If you start writing a message, and you just accept all of the things that it suggests for you - is it you, is it an echo of you, is it the AI?
50:00
LinkedIn does this in chat, right? LinkedIn and chat are very quick to make suggestions, like a bank, and usually, they're pretty bad suggestions. So, I don't feel like we're at any short-term risk of people just communicating through LinkedIn. But, OK, so...
50:17
That's kind of funny because I see those suggestions and I almost always put the other person's name in my reply to make it clear that I wrote it myself.
50:28
Do the same thing, you.
50:31
You have to signal to the market that it's you and you have to stay authentic to it. Yeah, I agree. I agree. Exactly. OK. Well, we're at, OK, this is always happening in these podcasts as we get towards the end and we're like, we could do yet another podcast from there.' So, Jeff, if it's OK, we'll bring you on later in the year and get reviews on what's top of mind for you then. But I hope you enjoyed spending time with us.
50:58
This has been a total blast. I, I, you know, of all the different ways to interact with audiences, these conversations I think are the best, you know. Certainly, there's the real formal way where you create a presentation, you stand up and you talk. many a lot of my recent live sessions, I'm saying, 'I could do that, but I'd much rather...'
I'd rather just have a conversation with you and with the audience; it's free.
51:19
Flowing, you know? Yeah, totally. I mean, we're big believers, so thank you so much for spending time with us. We'll bring you on later on again as we circle back on different topics and presenters. Best of luck in your journey.
51:35
Sounds awesome. I enjoyed being here.
51:39
Thank you for listening to Unlearn. Subscribe wherever you listen, and visit UnlearnedPodcasts.com for the transcripts.
.png)
.png)