Building External Evangelists
This is a modal window.
Das Video konnte nicht geladen werden, da entweder ein Server- oder Netzwerkfehler auftrat oder das Format nicht unterstützt wird.
Formale Metadaten
| Titel |
| |
| Untertitel |
| |
| Serientitel | ||
| Anzahl der Teile | 542 | |
| Autor | ||
| Mitwirkende | ||
| Lizenz | CC-Namensnennung 2.0 Belgien: Sie dürfen das Werk bzw. den Inhalt zu jedem legalen Zweck nutzen, verändern und in unveränderter oder veränderter Form vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. | |
| Identifikatoren | 10.5446/61804 (DOI) | |
| Herausgeber | ||
| Erscheinungsjahr | ||
| Sprache |
Inhaltliche Metadaten
| Fachgebiet | ||
| Genre | ||
| Abstract |
|
FOSDEM 2023541 / 542
2
5
10
14
15
16
22
24
27
29
31
36
43
48
56
63
74
78
83
87
89
95
96
99
104
106
107
117
119
121
122
125
126
128
130
132
134
135
136
141
143
146
148
152
155
157
159
161
165
166
168
170
173
176
180
181
185
191
194
196
197
198
199
206
207
209
210
211
212
216
219
220
227
228
229
231
232
233
236
250
252
256
258
260
263
264
267
271
273
275
276
278
282
286
292
293
298
299
300
302
312
316
321
322
324
339
341
342
343
344
351
352
354
355
356
357
359
369
370
372
373
376
378
379
380
382
383
387
390
394
395
401
405
406
410
411
413
415
416
421
426
430
437
438
440
441
443
444
445
446
448
449
450
451
458
464
468
472
475
476
479
481
493
494
498
499
502
509
513
516
517
520
522
524
525
531
534
535
537
538
541
00:00
MereologieOffene MengeZahlenbereichVorlesung/Konferenz
00:40
Twitter <Softwareplattform>E-MailOffene MengeOpen SourceHackerWeg <Topologie>Open SourceMetrisches SystemHackerSchreib-Lese-KopfProjektive EbeneComputeranimationVorlesung/Konferenz
01:10
MultiplikationsoperatorEinflussgrößeMetrisches SystemVorlesung/Konferenz
01:46
UnendlichkeitEreignishorizontWärmeausdehnungCodeAnwendungssoftwareWeb logRFIDGraphische BenutzeroberflächeTermProzess <Informatik>SoftwareProdukt <Mathematik>WhiteboardMetrisches SystemGemeinsamer SpeicherInterpretiererMinkowski-MetrikMereologieAussage <Mathematik>ZahlenbereichBitSoftwareentwicklerMultiplikationsoperatorVererbungshierarchieDifferenteWeb SiteEINKAUF <Programm>ComputeranimationVorlesung/Konferenz
06:56
RechenwerkKontinuumshypotheseSoftwareentwicklerMinimumMultiplikationsoperatorProduktraumBildschirmfensterBitrateMetrisches SystemCodeProgrammierspracheMereologiePaarvergleichSpeicherabzugDatenbankEinflussgrößeDifferenteKollaboration <Informatik>Produkt <Mathematik>RechenschieberVirtuelle MaschineGraphInhalt <Mathematik>CASE <Informatik>TermSoftwarewartungVideokonferenzEinsSchnittmengeYouTubeMinkowski-MetrikGoogolResultanteNichtlinearer OperatorProzess <Informatik>SichtenkonzeptTeilmengeVorlesung/Konferenz
13:53
HilfesystemVererbungshierarchieSichtenkonzeptProjektive EbeneGebäude <Mathematik>MereologieSprachsyntheseDifferenteMultiplikationsoperatorGemeinsamer SpeicherEinfach zusammenhängender RaumTypentheorieProzess <Informatik>SoftwarewartungMulti-Tier-ArchitekturPatch <Software>Exogene VariableBestimmtheitsmaßKlassische PhysikMetrisches SystemOpen SourceProdukt <Mathematik>PhysikalismusUmsetzung <Informatik>Vorlesung/Konferenz
20:50
E-MailOpen SourceProgrammierungSoundverarbeitungHilfesystemEndliche ModelltheorieAggregatzustandVererbungshierarchieCodeInformationErwartungswertMultiplikationsoperatorOffene MengeWhiteboardMereologieBEEPTypentheorieWeb logOrbit <Mathematik>Gebäude <Mathematik>MAPVorlesung/Konferenz
Transkript: Englisch(automatisch erzeugt)
00:00
Okay, and now is the part where we promise you, Matt. Go Matt! Alrighty, thank you everybody. Thank you. And Flora, you know, we're starting to warm up the crowd for you. You know, Flora was, said I was the opening act here, so I'm really, you know, glad to
00:20
be able to introduce the rest of the, you know, the folks today. But I wanted to talk to you about something that is really critical, it's building external evangelists. How many people are here in DevRel? A few, okay. Oh yeah, good number. Community? Just, you know, in general? Okay. Alright, well. Yeah, DevRel community. It's kind of, it kind of overlapped now, don't they?
00:40
So who am I? This is me. That's me over there. Yes. So I am the head of Open Source Strategy, or the HOSS for short, if you want to call me the HOSS, I can answer to that. You can find me everywhere on socials if you want to connect after the fact. I work for Scarf, if you're not familiar with Scarf, we do advanced metrics to track
01:02
adoption and growth of open source projects. And we also have the Hacking Open Source Business Podcast. So if you're interested in being a guest or checking it out, appreciate it. So one of the questions that I get asked a lot, and I talk with founders, I talk with executives quite a bit, I get this question asked all the time. How do you measure community or DevRel?
01:22
Has anybody gotten that question in the, you know, the space? In the last five minutes, just yesterday, right? Executives have a hard time figuring out, you know, what this is. And my answer generally to them is, you measure the external advocates or evangelists you
01:40
create. Okay? There's a lot of other metrics you can track, but this is the one that I typically go to more often than not. And you might be asking the question, why? And so I'm going to answer that question in 348 slides. I think I have until noon today, so I should be going for a while here. That's cool. But my first question on the why is, what is the company expecting from these activities?
02:03
So when you're talking about community DevRel, what are you expecting? You know, so anybody want to shout out what you're expecting? No? Okay. I'll just go. I'll tell you what most people are expecting. They're expecting more contributors or more users or more customers. And what's funny is, every company that I talk to, they might fit into one of these,
02:23
but they want all of them, right? So if they're like, oh, we want more contributors, well, really, what do you want? Well, we want more people to use our stuff, and if we think we have more contributors, we'll have better software, and then if we have more users, we'll have better, you know, customers, right? So this more is something that people are striving for.
02:41
So no matter what, people are looking for more. So I have to put on my other hat, because those who are DevRel in the community space, I got to go a little sales marketing business-y, so I got to switch hats. I got to put on my sales hat, which is the super villain hat, so I can talk a little bit about other things other than community activities.
03:03
So most companies are looking for more dollars, okay? Whether you're looking for more contributors or more users or more customers, at the board level, at the executive level, ultimately they're thinking dollars and cents. And it gets back to a hypothesis, right, which is the more users in the community, the bigger
03:25
the community we have, the more potential customers we have in the future, right? Have you ever seen the product growth flywheel before? It's kind of overdone. This is my really crappy interpretation of trying to redo it, so if anybody wants to redo it in a prettier graphic, you can see that.
03:45
But the idea is the more people who come in, find your software, try it, start to use it, start to be part of the community, the more willing they are to either tell other people about it or potentially purchase something for you. And this is where you get a huge disconnect, right?
04:03
So you have the community dev rel team here who is trying to be very focused on the relationships that you have with your community. You're trying to track the community growth, and you're not necessarily thinking about the number of customers, ARR, MRR. But when you get to the boardroom, a lot of times those are the metrics that they're looking
04:24
at, whereas we're talking about these metrics. And so there's a disconnect, and we need to link those two somehow. And early on, a lot of these are talked about quite frequently in smaller companies, but as you get bigger, everything tends to shift to this side.
04:41
And when you're saying, ah, how many of the number of customers or the ARR can we attribute back to this side, it becomes kind of squishy, right? And that becomes a difficult proposition for a lot of folks. And this is where we've seen the rise of the quote unquote influencer marketing or dev rel in our space, but influencer marketing has been around for a while.
05:03
Maybe you've seen some of the documentaries, or you've seen some people out there on Instagram who are like, look, I have free trips, and I get to go to all these places. You know, hey, I get free trips to go to these places, too, and stand in front of you awesome people. So I'm kind of an influencer, right? But this is trust, right?
05:22
It's all based on trust. So when you see somebody, whether it's online, whether it's part of some other community, you start to get to know them, and you start to trust what they say. And when they tell you to buy a product or do something else, hey, that's something that you listen to. And that has been established in not just the software space, but also in the overall
05:45
kind of retail and consumer space as well. And as tech evangelists, we are all part of that, even if we don't want to consider ourselves, we are. We are influencers. Our job is to go out there, talk to people, and get them to understand what we're trying to preach to them, tell them, share our experiences, help them get better.
06:07
But there's a difference between internal and external evangelists, right? So let me give you this example. We've got two people here, right? The first person works for Delta Company, let's say, and they do development JavaScript. And they say Delta's software is awesome, OK?
06:23
And B works for a website that uses Delta software, and they say Delta's software is awesome. Who are you more likely to believe when they say that? You're going to believe B. Why? Well, this person over here is incented to tell you that Delta's software is awesome.
06:41
So if I come up here and I work for Scarf and I tell you, Scarf's stuff is awesome, you'll be like, oh, OK, that's cool. But if someone else comes along and says, Scarf's software is awesome, you're more likely to listen. And that's why building these external evangelists is critically important. And from a business perspective, this is how the business side thinks of this.
07:05
Someone out in this space tells you, this is great, this is awesome. Their followers, some of these blue people here, they're like, ooh, yeah, that sounds pretty cool. Let me go ahead and let me watch this thing that you're doing or attend a conference or listen in.
07:21
And so those people watch. And then some subset of those will say, ooh, I really want to try that out. And so then they try it out. And of those who try it out, maybe a couple will actually adopt it. They'll maybe become customers. And that funnel, that process, is what is expected from the business side.
07:41
And that's what the investment in the DevRel space is all about when it comes to the executive side of things. But the thing is, can this scale? So if there is one me, so I can go to one conference at a time. If we hire another me or another me, we can do this as well.
08:01
But that gets really expensive. And a lot of companies, especially in the younger stages, are a little strapped. So they might have one, two, three DevRels, maybe. And so how do you scale that? And that comes into thinking about this in terms of different tribes or different communities.
08:22
There are lots of different programming languages. And it's not feasible for one person to understand all the different tech ecosystems, all the different tribes, all the different communities that are out there. They're all very different. And so by plugging into that and finding people who might be able to plug into those external communities, those external tribes, you can actually do this play where you're talking
08:45
to people and getting them to try over and over again in multiple communities that you might not be part of or might not be able to reach. So how do you get external evangelists? There's a few different ways. This is not an exhaustive measure by any means.
09:01
It's going to be the 15-minute version, which is the time we've got left. But let me start with, should you hire influencers or DevRel people or external people to come in and do something for you? There's a lot of people who have a really high GitHub rating. They have a lot of YouTube subscribers.
09:21
These are great people. But your results are going to vary because their communities might not match what you're trying to do. And so I've done this in the past at previous jobs where you partner with someone, and generally it's a paid relationship. And you're like, hey, could you cover this event? Could you come talk? Could you do these things? And sometimes it works, and sometimes it doesn't. It's a very costly way of doing it, but it can be effective if you do it right.
09:45
So the classic play is, obviously, grow your user base. That's easy, right? So OK, we're done. Just grow your user base, you're done. It goes a little bit beyond that. So when we talk about growing the user base, this is where you look at the graph here.
10:02
You start with users, and there's different pathways for those users, but everyone has to start as a user. It could go into the dev side where they want to be code contributors and eventually become maintainers or part of the core teams, or it could be that they become content contributors. Someone who does blogs, someone who talks at conferences, and then eventually turning
10:23
into a regular evangelist who continually does that. Now what's interesting is a lot of folks, especially in the community, a lot of our metrics are based on this top, which is great, but a lot of this bottom is not necessarily looked at or followed as closely, and there isn't as flushed out a metric set for these.
10:43
So how can you get more users? There's a lot of different ways, right? So content, content, content. Content machines. I do a lot of blogs, podcasts, videos, all kinds of different stuff. That helps people, especially if it's geared towards making it easier for them to get into
11:01
the community. So education, training, code examples, conference talks, things like that. But also there's a product space here, features that matter. That's important. So it is a collaborative effort with product in a lot of cases. You have to get that. You have to make it easier for people to get started. And then you have to figure out how to get more people to tell how awesome your product
11:25
is. That's the whole evangelist side, right? Because again, if I tell you we're awesome, it's okay, but if someone else who you trust in the community does, who has nothing to do with us, that's even better. And you have to set those expectations, right?
11:42
And this is a classic problem that I see happen over and over again in the startup space, especially in DevRel and community, is you start to compare people to other people, okay? Not all communities are the same, and you have to identify that. So if you're part of, let's say, the Haskell community, and you're gonna compare with JavaScript,
12:04
those are two vastly different sized communities. And if you're coming from the JavaScript community and you're saying like, look at all these people in my conference talk, I've got a full room, I've got 200 people, you go to the Haskell and it's got 30, you can't be like, oh, I only got 30 people at my talk.
12:21
That's not what it's about, because the Haskell programming language just has a smaller community, and that's okay. I'm a database guy, and when you look at databases, for instance, I built this chart, it's really hard to read, but you can get the slides after. But the gist of this is, you start off and you think, wow, how many people use databases?
12:41
Everybody uses a database, so what's your total market size? How many people can we get to use our stuff? Oh, there's billions and billions of people who are gonna care about what I want to say. Well, when you start talking about operationally sides of things, you're like, huh, okay, well, from a developer perspective, let's be honest, four million developers don't care about databases,
13:01
so half of all developers go out the window. And then you're like, hmm, well, maybe full stack ones, I guess, care about it, but architects do? And then you're like, oh, well, what about operations people? Well, DBAs do. Do sysadmins? No. So you have to start to pare down and figure out what that audience is, and don't get discouraged
13:21
if you're working in a space where the audience is small and you have to grow it. That's something that I see happen quite often, and a lot of executives and companies, they're like, oh, that guy got 10,000 views on his video, why can't you do that? It's like, because I'm not them. I had an interview with, I talked with a guy from Google on a podcast that I did, and he's
13:41
like, he's in the database space. It's like, the Google guys, they don't get it, because our Google videos on G Cloud, they're like 500,000 views, and the database ones are like 5,000, and they're constantly harassing me. You're not doing enough, because you're not getting the 500,000 views. Well, the community's smaller, it's different, and that's OK.
14:04
So avoid some common mistakes, right? Don't assume that your audience is as large as others, and don't assume that just because you're speaking will automatically generate interest. It requires a lot of work, and product awesomeness trumps story.
14:23
You can have the best storytellers, the best DevRel folks out there, and if they're talking to people about a crappy product, it just doesn't work. So some best practices. Be authentic. This is really important. You want to connect with people. You have to be authentic in your conversations.
14:41
If you are, you know, authentic people will come to you, and you have to make connections. How many people like dogs? OK, half of you. So you know what? This is my dog. And you know what's awesome about dogs? You can be an introvert, all right? And if you're walking around, you know, no one else in the neighborhood, if you have
15:00
a dog on a leash, they will stop by and talk to you. Why? Because dogs are cute! They also have dogs, especially if you have two dog people. I've met so many people just from the dog, you know, that it's not funny. It's a connection that you make, so how do you find those connections in the community? Because the more of these types of connections, the more people that you can get to potentially
15:25
help within the project, and potentially talk about your stuff. You also need to make this really accessible. People often develop brilliant things that are too complicated.
15:40
We also have maintainers and devs who think that everyone thinks like themselves. If you are comfortable on the command line, and you are comfortable setting up something following a 142-step process, don't assume that other people do that, right? You know, not everyone thinks that way.
16:02
And also make it accessible for how new users get involved. Be safe in a fun environment, right? Nobody likes to come and listen to the person who talks in the monotone voice, and is, this is a really boring talk in a boring community. You want to make things fun and engaging for people, right?
16:21
You want to connect with folks as much as you can. Avoid toxic people because they suck, you know? Avoid abrasive culture! What are you doing to engage with the community on a regular basis? What's the personality of your community? Does your community have a personality? It should.
16:40
You should define that up front. You should figure out what you want people to know your community as. You also want to be transparent, okay? Share what's happening internally. Oftentimes, especially in an open source driven company when you have commercial open source, you have gotten different tiers, right? You have the tier of people who are in the company and they know everything that's going
17:02
on, and then you want all these contributors outside to just assume that they know what's going on internally or not know it. And setting up those silos is bad. So sharing plans, roadmaps, other things, super critical. Don't create that caste-based system, okay? Explain yourself and your decisions, right?
17:21
When you say, I'm not going to take this feature, that's okay. You can do that as a maintainer. But if you don't explain why, then that really hurts people, right? And also be open to feedback. Now the next one is, you really want to focus on getting help, but you want to ask for it.
17:44
I don't know why so many people are afraid to ask for help. This is a classic blunder. If you're running a community and you want to build contributors, maintainers, you want to build external evangelists, you have to tell people where to start.
18:00
Classic. I've seen this several times where people are like, I really want to get involved in this community. So they submit some patch and it's like, wow, this has nothing to do with what we want in this project, so we just reject it and move on. And then that person moves away and they never want to come back. Well, why did they choose that? Well, they didn't know where else to start. Tell them where to start. Help them with that.
18:20
And give that feedback, right? Be responsive. Something drives people more bad even, not hearing from you when they ask you a question or submit something to a project. It's mind-willingly bad. So you're building relationships here. Treat each other like you want to be in a relationship with one another.
18:42
You want to be part of this community. Avoid that silence. Good, bad. And promote people's work. So if somebody submits a blog, somebody's had a conference talk, go ahead and promote it for them. Tell them that this is awesome. Give them feedback.
19:02
Keep on rewarding people. People are rewarded by positive achievement. And that doesn't just mean baked goods. I do appreciate the baked goods, though. But you give away challenge coins, Tom. So there you go. Lots of different things you can do. But one reward that's often overlooked is simple acknowledgement.
19:24
I acknowledge that you contributed. Postgres does a great job of this. They give away challenge coins as well. But everyone who contributes, they're out in their read-mes. And every time they do release notes, right? So make sure when they do release notes or blogs, you write profiles on people.
19:42
Physical rewards, of course, work too. I like hats. But that's just me. But the other thing you can do is listen. Listening is critical to all of this. And really, just don't be an ass. People are out here to participate in the community.
20:03
And they want to be part of this. And so the more that you can help them feel like they're part of it, that means that they're going to be wanting to stick around and help more. External evangelists aren't just participants, right?
20:22
That's what you want. You want to have the external evangelists. You don't want people just in the community. You want to make sure that they can contribute and be happy to do so. So if you're interested in more, because I'm just about out of time, go ahead and follow along. Get more details. We just released this new website, opensourcemetrics.org.
20:45
And so I put all of the metrics that kind of go between the community and the business, and how to map out that entire funnel. And you can also see me later on today where I give the Open Source Business Guidebook
21:01
over in Room K. Or if you're going to the State of Open Con in the UK, I do have a talk on how to destroy a community, the Super villains guide. So if you really want to know how Darth Vader, Emperor Palpatine, Voldemort would go about destroying an open source community, stop on by and see me.
21:20
And again, that's where you can reach me. Thank you, Matt. Yes. Thank you very much, Matt, for that wonderful talk. I was actually just posting that anyone who's looking to get into a DevRel career, there's some really great information here about setting reasonable expectations with your
21:40
company leadership. So very much appreciate it. Matt, if you would take questions, we have a little time for questions. Now is the time to ask questions. Is it on now? Yes. No. Mic on. Mic on. Mic off. Okay. I can go.
22:01
Questions. Questions for people. Anybody have any questions? Anybody want to see me do stand up comedy? Yes. Yes. Yes. Yes. I'm embarrassed. I can't do it. Question.
22:20
Hold on a minute. Oh, no, no, no, no, no. I shall slowly walk through the audience. I really appreciated the advice to listen, to hear what people are saying so you can kind of bring them into their community, but how do you get people to create content about
22:42
something? Like, do you reward them, encourage them, is it just bragging rights? What's your approach? So I think that there's a couple of things that I've seen work effectively. I know there's a talk this afternoon on building non-code contributors, which is good, so I'd recommend that.
23:01
But typically my approach has been to help people get involved, start to work with them, talk to them initially, especially a lot of people who are just getting started. They don't know what to write. Especially when English isn't their first language and you're going to post in English. So working with someone, giving them ideas, I generally will start with a creative kind
23:20
of idea board and do that help-wide, where I'm looking for a topic on how to do X in Kubernetes or how to deploy this or how to tune this, and if they'll agree to take one, I'll work with them to release it, we'll promote it, so there is some of that reward. But I also go through and do monthly rewards for the top blog posts and contributors that
23:42
way. I'll look at yearly rewards for that as well. We've done things where given plaques out at conferences and stuff for someone who's had the most effective or interesting blog, had them come out to the conference, stand up on stage, give a talk, whatever, and that's a reward as well for them.
24:01
But a lot of it is just building that relationship and working with them, especially as they're starting out, because when you're starting out, you're really unsure and there's a lot of that imposter syndrome, you're like I don't really know if this is good enough to publish. And the thing is, you've just got to make it so easy for them to get started and feel like they're part of the community and feel like no matter what, you're there for them
24:24
and you can help them through it. Flora has you. Oh, sorry, thank you. Yes. My phone started to beep and my brain immediately turned off. Thank you, Flora. Hey, Matt, thank you.
24:42
So there's a couple of companies that have these community ambassador type programs. What in your experience or from your viewpoint are some good ones and why? So the community ambassador programs, like GitHub has the GitHub stars program where they've got quite a few people who are on the GitHub side of things, most of them end
25:01
up being more on the code contributor side and they don't necessarily kind of embrace the bigger audiences. Even when you look at a lot of the tools that are out on community space, whether you're talking about orbits or common room, they're very focused on the contributor first and then a few extra things, but they don't necessarily do the full picture.
25:21
And so I think that that's where if we want to continue to grow the community, it has to be in embracing all kinds of contributions, not just the code contributions. And so I think that that's where we need to kind of move to that next level. And when I was at Percona, we tried to build that program in, but it was still relatively
25:41
young. So we were still working through the kinks. We went through one year on that program to do that, which worked pretty well, but I wouldn't say that it is the model for everything just yet. Thank you very much, Matt. I don't, I, oh, it is on. I love it. Thank you.
26:00
Thank you very much for your top floor. If you would come up please and get set up, that would be lovely.