Case Study: Skool

0:00

Embedding Native Video makes it a lot easier for the communities to share and

0:05

grow and learn

0:06

together and it protects the content for these people that are investing a lot

0:11

of their time

0:11

and energy to build their communities on school. We want to be the best

0:15

platform for the most

0:16

people and for the least price so Native Video is available in both plans for

0:21

free for all of

0:22

the creators and all of the users inside of the platform. We're taking the cost

0:27

on ourselves and

0:28

muxes allowed us to do that. Hi, I'm Matt McClure. I'm one of the co-founders m

0:38

ux. I'm thrilled

0:40

to be here with Kirby and Andrew from one of our new favorite customers, School

0:46

. Real quick,

0:47

Andrew, you want to show us your shirt? I'll show my shirt. Video nerd, video

0:52

nerds represent.

0:54

You said you're you're brought in video natively. Can you walk me through like

0:58

what were the

0:59

what were the video challenges you were trying to solve or her like what did it

1:03

look like before

1:04

before us came to the party? Yeah, absolutely. So previously and still now

1:09

users would upload

1:11

videos to YouTube or Vimeo, Bloom, you know, a variety, whatever they wanted

1:16

really and then

1:17

embed that into school for their courses or if they wanted to do a post or your

1:22

comment

1:23

with a video link. And so that's there's a pretty high barrier there of like,

1:28

well,

1:28

I've got to record my video. Now I've got to go to another platform, upload

1:32

that, wait around for a

1:34

while for that to finish processing. And I've got to make sure that it is like

1:38

publicly available

1:39

but also de-listed because I want people to go to my school community to view

1:43

this and then I link

1:44

it there. And so there's like multi-step process and really we just want to

1:48

lower that that burden

1:50

as like so many platforms and we want school to be a singular place for this

1:55

community.

1:56

And so that's where a native video comes in for group owners or admins to

2:00

create their courses,

2:01

they can drag them right into school. And now you can only view them on school

2:05

as opposed to a

2:06

YouTube link that somebody could just, you know, network inspector or you copy

2:11

the YouTube link

2:12

and then share it anywhere else. There was a post that we made when it was

2:16

announced and the amount

2:17

of fire emojis, the amount of people being thrilled was insane. You know, we've

2:21

had in that post some

2:23

some community leaders with really large communities are like, well, I'm going

2:26

to save $1,000 or $1,500

2:28

a month on my video hosting. But by just moving over to school native video, we

2:33

've seen a lot of

2:34

you know, improved engagement in posts from the community members, skate life,

2:39

professional skate

2:40

boarder, helping the community, the skate community grow together, you know,

2:44

and so he's like, hey,

2:45

try a 180 variable with heel flip. I think that's right. And now rather than

2:50

having to record

2:51

yourself uploaded, you know, you just record it, bam, right in there and you

2:55

can get feedback

2:55

from the community, from the leaders, just instantaneously. Why, why choose MUX

3:01

?

3:01

What was that like, what was that like light bulb moment for you to realize

3:05

that that MUX could

3:06

be a fit or solve these problems? Yeah, schools of startup, right? So, so, you

3:11

know, we're not a,

3:12

we're not a multinational conglomerate. We have very small engineering teams

3:16

with very

3:17

full roadmap. So looking at everything we want to do and the scope of effort of

3:22

building a,

3:23

you know, ingest pipeline, storage platform, delivery, all of that stuff is

3:28

like,

3:28

we should consider other options. And I've been doing video for a long time. I

3:33

've used MUX data

3:35

for four or five years at previous, you know, jobs and you just really love

3:41

that platform

3:42

familiar enough with the API. And so it wasn't really a light bulb moment of

3:46

like,

3:46

maybe MUX. It was kind of, MUX was, for us, it was the starting point. Once we

3:51

said, hey,

3:52

we're not going to build this fully ourselves right now. We started with MUX.

3:57

We did our due

3:57

diligence of looking at other platforms. We looked at the, the technical

4:02

performance,

4:03

the feature set, the finances, and MUX was just number one in all of them. And

4:08

so it was a really

4:09

easy decision to make. Let's go with MUX. And, you know, we were able to build

4:14

with that leverage

4:15

of the excellent APIs and the features that we need. I think it was from saying

4:19

, yeah,

4:20

we're going to do it with MUX. It was about five weeks to going lies. And the

4:24

confidence that

4:24

we had in the product, we didn't even do a slow rollout. We said, on our, on

4:29

our weekly broadcast,

4:31

hey, native live is here. And we flipped the switch in 100% deployment.

4:35

So, and it worked, you know, really, really low amount of issues and a lot of

4:42

happy users.

4:43

That's music to my ears. Any specific MUX features or capabilities that

4:51

stood out? Like, you know, I always love hearing about like interesting use

4:57

cases of

4:58

thumbnails and workflows or whatever else, but anything in there specifically

5:02

that you were

5:03

like, ah, man, cool. Yeah. So I'll, you know, a technical side for me is the

5:10

upload component,

5:11

you know, the multi part upload up chunk, I think you guys call it. So just

5:15

being able to

5:15

dry it, like drop that component right in. And we don't have to deal with

5:19

multiple signed URLs

5:21

and stitching those things back together. All that, you know, just here's the

5:25

thing and it works.

5:27

And it's fast, right? So, so the technical side is that the thing that like

5:34

stands out for me,

5:34

though, more than that is the people at MUX. I've been doing video for a long

5:38

time. I have a pretty

5:39

high bar. And I'm not shy about needing that bar to be met. And so I've been a

5:45

little bit noisy

5:46

about some of the things as we come up on hurdles, some of them us, but your

5:50

video is hard. So there

5:51

are going to be issues our users are doing their own thing. So every step of

5:56

the way,

5:57

regardless of how annoying I might be, everybody at MUX has just been a

6:01

pleasure to work with. And

6:03

so that's the real thing. It's, it's, you've kind of solved the video as hard

6:07

problem. And now it's

6:08

like the, but the people are great. It's also good to hear the, the shout out

6:13

to upchunk. I actually,

6:14

I wrote that years ago, initially, what would be, I mean, outside of just like,

6:21

you know, your,

6:23

your startup. So not having to build out like a big video infrastructure team

6:28

is, I assume,

6:28

that's a big win. But like, I don't know what would be, what would be

6:32

impossible or a lot more

6:33

difficult, if, if MUX wasn't in the picture. I mean, literally everything, like

6:39

, we, we went from

6:41

concept to full production 100 deployment in about five weeks, like, we're just

6:48

not going to be able

6:48

to do that. The uploader itself is going to be probably that amount of time

6:53

just to implement,

6:55

right? And so, so there's that just the, the baseline feature set is so

7:00

comprehensive. We

7:01

didn't really have to worry about any of the things that, you know, thumbnail

7:05

creation, the

7:06

hover scrub sprite sheet, all of that, the MUX player integrated right into MUX

7:11

data. So we have

7:12

analytics and we know we're getting our, you know, engagement, so that we can

7:17

see, hey, actually,

7:20

on posts, it's about 98% of videos get views, which is for UGC is an insane

7:27

rate. So that's,

7:29

that's huge for us to be able to figure that out, right? And so all of those

7:35

things on the

7:35

technical side are just impossible to do without a partner like MUX. Another

7:41

thing on the members

7:42

side is that school is a community platform. So we have a community for our

7:47

community owners.

7:48

And sometimes you can say things that you later regret when people ask, when is

7:52

native video going

7:53

to be out and you tell them, ah, it's going to be out soon. And then when it

7:56

comes to actually

7:57

building it, the engineers are like, this is not going to be out soon. If we

8:00

were to build it

8:01

ourselves, all of those promises that we had made to customers that it was

8:04

going to be out soon

8:05

wouldn't have been met. So having a partner like MUX allowed us to get it out

8:08

so much quicker

8:09

to the point that customers were like, how did you build that so quickly? I

8:12

have no idea how

8:13

we've got native videos so soon. So that was really cool as well. Yeah, if we

8:17

were to go and build

8:18

fully true native on AWS, we'd have to do like five year long commits with

8:26

these growth plans.

8:27

And you know, we're just not, we're just not there yet. So leveraging MUX's

8:31

scale

8:32

is part of what has allowed us to give it away to every user for free as just

8:37

part of this is

8:39

right. It's a baseline feature of school. And there's other competitors that

8:44

charge the creators

8:45

for native video. Yeah. And there's also competitors that charge the members to

8:49

upload longer videos.

8:51

And we didn't want to do either of those things. Yeah. The 98% of people, the

8:57

98% of video upload

8:59

being viewed at all is kind of a crazy stat. Is there any statistic around like

9:06

uptick of native video, like how many of your communities are utilizing video

9:12

in some way?

9:12

Thousands, thousands of communities are engaging with this. It is very quickly

9:20

become

9:21

the predominant usage of video. Folks are still using YouTube or Vimeo or the

9:27

others

9:27

because that's part of their flow. And we've seen a very quick uptick in usage.

9:33

Really,

9:34

it's an exciting new addition in the platform. But those that have their flow

9:38

are starting to

9:38

peel off and say, I'm going to stop sending my video over there. I'm going to

9:42

start bringing it in.

9:43

And so a large initial influx and then just steady growth for groups and

9:51

members. And

9:52

as each group kind of starts getting used to it and understanding how it works,

9:57

we're seeing them start to encourage their community say like, hey, don't do

10:01

the old flow.

10:02

Like drag and drop your stuff right in. It's way better. And so we're seeing

10:05

those community

10:06

leads telling their community members use the school video. It's really good.

10:11

Right. It was particularly fun. The first like two days of turning this live

10:18

for everybody was

10:19

the videos were coming in like five to 15 seconds long. It's like, that is not

10:25

what I expected.

10:27

But then then it equalized and you started moving courses and posts and stuff

10:32

over to

10:32

native and we saw those go up. And what we realized is that was just everybody

10:36

being

10:36

excited, taking a quick video, you're like, Oh my God, it's here and uploading

10:40

it. And so they were

10:41

just, it was a lot of real quick phone videos. Oh, is it does it work? And then

10:46

they'd be excited.

10:47

And then over the next several days, they start moving their content.

10:50

And yeah, I mean, I know it's fresh, but any, any initial signs around the

10:56

impact, like are you

10:58

seeing more folks, you know, for the communities that have started using this

11:02

and utilizing it

11:03

more, are you seeing like more engagement there? Like, can you point to any

11:06

positive signs for

11:09

business after it's yeah, it's kind of tough because we, we also launched the $

11:14

9 hobby plan,

11:16

the next week. And so that, that little bit of lag of like, it took a couple of

11:20

days for folks to,

11:22

to get used to and start using it. And then we introduced hobby plan. And so

11:26

yeah, our graphs are

11:27

going way up as they go to the right. Let's just say it was Mux. It's both of

11:31

them. It's definitely

11:32

Mux on the $9 hobby plan, I think is the killer combo. Yeah, right. And even

11:38

not from a statistics

11:39

point of view, I know lots of creators personally that have moved across

11:43

because of native video.

11:44

I have one friend who has a community of guitar players, he helps people get

11:48

better at guitar.

11:50

And he was just like, for years, probably four, five years, he was like, I can

11:53

't move across,

11:54

I can't move across, because everyone uploads videos and I review them, getting

11:57

them to use

11:58

YouTube or Vimeo or Wisti or Loom, it's too complicated. And as soon as we

12:01

released it,

12:02

he moved across. So that's just one guy, but I'm sure if there's one guy that's

12:06

lots of other people.

12:07

So yeah, are you guys already using it? You know, transcripts, captions, things

12:12

like that.

12:14

You know, either as part of these workflows, but also just accessibility, like

12:18

general

12:18

accessibility, like how are you thinking about that in terms of the video side

12:21

of your communities?

12:22

Yeah, definitely. We are using it on every video right now. And that's

12:28

something that we're

12:28

looking at maybe given the community members an option to turn that off or

12:33

select the languages.

12:34

Again, we kind of built the minimum viable product that actually just hit all

12:39

of the core features

12:40

that we needed. But with the language options that Mux provides, we're going to

12:45

look at,

12:45

you know, making that selectable by the group. We want to add in the

12:49

functionality to do multiple

12:51

language transcriptions. You know, if there's multiple audio tracks or if there

12:56

's, you know,

12:57

if Kirby, we're speaking French and I'm English, you know, we go back and forth

13:00

, we want to have

13:00

both options. But we're definitely using those and the community at large is

13:04

really appreciative.

13:05

Because now you can you can watch your school courses, you can watch the

13:10

content on the bus,

13:11

right? Or on the train and read along. You know, if it's too noisy or you don't

13:17

have headphones,

13:18

it just provides you another layer, another option. If you were recommending M

13:24

ux to another

13:25

dev, another dev thinking about building something, like facing similar video

13:29

challenges, whatever

13:30

else. What would what would you tell them? Don't do it yourself from the start.

13:34

You know,

13:35

hosting video at scale at broad scale is a lot more than just wrapping FFM peg.

13:44

I mentioned the

13:45

uploader itself. That's a fair amount of work. And for that to be proven and

13:49

reliable and battle

13:50

tested for years, right? Is something that you you don't just get when you whip

13:55

out some code and

13:56

deploy it. We are we are growing. We have a lot of video coming in, we're

14:00

getting a lot of views

14:01

and the views are continuing to grow. And if we were just doing it ourselves,

14:05

we'd be paying retail

14:06

rates for CDN, right? Because we don't want to make that five year commit,

14:12

because we don't we

14:12

don't know we're still learning as we grow. And Mux has done all of that work,

14:18

right? Multi platform,

14:21

multi CDN delivery at the lowest CDN rates. I think you guys also have your own

14:26

CDN to some extent.

14:28

And so that's all stuff that you don't want to do with a small team, because

14:33

you'll just be

14:33

underwater instantly. So at the very least, start with Mux, use Mux data,

14:39

figure out how your product

14:42

is being used. And then if you're really insistent, then you can start to peel

14:45

away certain parts,

14:47

you know, as you grow, or you know, if you want to go build your own CDN, go

14:52

for it. I recommend

14:55

against that. But yeah, start, start with Mux, and then see what you can do

15:02

after that.

15:03

I think the core thing, the theme for school is community. I think the Mux, you

15:11

know, is a big part

15:12

of the video engineering community. And so it just felt really natural to to

15:17

blend the you have to

15:18

become partners. You guys are not like a faceless corporation. And as I said,

15:24

the people are just

15:25

great to work with. And as people that I've known for a long time, because you

15:29

're you're leading a

15:30

lot of the effort for making video doable, right, and and lowering those

15:34

barriers. And so yeah,

15:38

the school community aspect, the Mux community, D Mux, all of the other things,

15:42

was just added

15:45

to that natural fit. Thank you all so much. Really appreciate it. But thanks

15:51

for the time,

15:51

words, and thanks for doing this. Of course, yeah, I'm thrilled you guys are.