Example Product Spec

[00:00:00]
>> Spoiler alert, I'm gonna ask you to write a product spec here in just a second. However, I thought it was unfair for me to not just show you an example first. So I'm actually gonna show you the answer first, right, and then I'm gonna ask you the question.

[00:00:15]
So I came up with a random product idea while I was sitting there on the couch. Let's pretend for a moment you're a PM at Netflix and your task is to add a feature that allows users to video chat like Zoom or FaceTime with someone and watch a synchronized stream of a show together.

[00:00:31]
So imagine you're doing a long distance relationship, you're in Seattle and she's in New York or something like that. So you could watch your show together with your significant other or something like that. Or maybe your friend wants to watch you watch love, death and robots, cuz that show`s weird and it elicits feelings from people, and don't watch it unless you're an adult [LAUGH].

[00:00:55]
So, that would be another reason that someone might wanna watch that with someone together. Is that a good idea? I don't know. But let's say we put together a product spec to then go say maybe we wanna try this, maybe we don't wanna try this. Here's how I would write that spec.

[00:01:12]
For the record by the way, I don't think this is a good idea, but it was one that was easy to produce a product spec for. So I wrote this last but Netflix users frequently express a desire to watch shows with friends and family remotely while engaging in real-time discussions.

[00:01:31]
Previous attempts to implement such a feature have not met the desired results. The goal is to provide users with the ability to watch Netflix content together, regardless of their devices, and have a synchronous video chat similar to FaceTime. Does that describe the project to you more or less?

[00:01:50]
ChatGPT, I don't think I even modified that one. That one was actually quite pleased with that came out with, so I just stuck with that one. So problem statement, this is blown out a little bit. I actually put the user story directly right there, I'll frequently do that.

[00:02:06]
I talked a little bit about here that Netflix has tried this before and it has been met with varied success here, but we still get requests for it. So there's still a problem here that users are looking to solve, here's another attempt at it. Where I can put statistics in here, I'll usually do that as, hey, this has come up this many times during support tickets or X% of requests from this.

[00:02:31]
And so here's our proposal to solve said problem. Goals, allow two people to watch a show together, regardless of the device type as long as it has audio or video. We make no attempt to shows things that like it's available in one region and not in another. We do wanna allow users to pause, scrub is what you do when you move back and forward in time in a video, that thing is called the scrubber.

[00:03:02]
Have it work for both users at the same time and allow users to use their own subs and dubs. So, if I'm hanging out with my Japanese friend and they wanna watch in Japanese and I want to watch in English, that's okay. Non-goals, cuz you can imagine someone's gonna raise their hand in the back of the meeting.

[00:03:18]
What if I wanna do it with three of us, right? This allows us to explicitly say, not for this version. If this is wildly successful v2 of this sure, let's do three users. We're not trying to sell for anyone that doesn't have video device, we're not making webcams, we're not going to do anything of that nature.

[00:03:36]
And we're not gonna do anything to encourage users to chat without watching a video, right? So we're not gonna allow them to just use Netflix as FaceTime essentially, okay? Key metrics, time watching Netflix using Watch Together, time spent in chat total. We wanna check specifically, is this driving referrals of Netflix, right?

[00:03:59]
So if I sign up for Netflix and then immediately use this feature, it probably means that other person influenced me to try it and do it so that we could watch it together, right? Like if I have it and maybe my girlfriend doesn't have it, right? I probably got her to do it so that we could watch it together or something like that.

[00:04:18]
Month-over-month on retention. I imagine a lot of users are gonna try this and then not try it again. That's a very common pattern of usage, so I wanna make sure this people are finding enough use in this that they continue to use it. And then are they satisfied with this?

[00:04:31]
We'll give them a little post surveys like, was this feature useful to you or did you have problems blah, blah? This is obviously a big experiment so rollback criteria here is, after three months we'll look at initial metrics if we can pivot. After six months we're gonna expect to see and we'll look at it again, six months and then if by nine months we don't see any sort of growth during that time, we will begin with winding down the feature.

[00:04:56]
Netflix is really good about this. I'm sure a lot of you have seen Netflix will try things and then spin it down, right? Trying to think if there's any good examples that I can point out to. Skip Intro was experiment that we ran while I was there and we had no idea that people would actually use it frequently.

[00:05:15]
Cuz we generally had a hypothesis once people click play they don't really care what happens. Turns out that was very not true, it's extremely popular, right? So that was one of them we found surprising, and it would have been rolled back if people hadn't used it. Timelines, product specking period, this red dot is where we are today.

[00:05:35]
Technical architecture design working from the same time development. We'll have a product review here, a beta release here and general availability here, if it's works out. And then I pulled out the date specifically because cuz that's something that executives want to see. Okay, and here's my Mock UX, right?

[00:05:52]
I literally just went on my phone and just took screenshots and drew over with MacOS Preview. So, user opens their app and with the show that they wanna watch, they click on watch together you can see here I put a little emoji there with some text on it, right?

[00:06:10]
You can see as that this color doesn't match, this is definitely not centered, who cares, right? This communicates my point, perfectly. Little modal shows up. This one already exists in Netflix, I didn't even have to redesign that at all, right? User shares the link with their friend and then they end up on watching something up here.

[00:06:32]
You can tell this is obviously not the correct video, but you know what I'm talking about and I'm not gonna spend a bunch of time trying to make this look anything else. As you can see, this probably took me what all of 30 seconds to draw some boxes and put some emojis on a page.

[00:06:46]
And then I took a stock photo and put it right there. Done, but you know exactly what I'm talking about, right? So caveat there, that's a stock photo and got it from Unsplash and the designs belong to Netflix with only minor edits for me. That's what I'm talking about with these super low fidelity super hacky kind of designs.

[00:07:08]
This is what I want you to do. I want you to be crafty about it. I want you to be fast. I don't want you to spend a bunch of time on this. Think if I had gone into Photoshop and spent a bunch of time on this, I would have spent, I don't know 10 extra time, 20 extra time and how much more effective would have that been in my spec?

[00:07:30]
I'm gonna say exactly zero percent more effective, right? Maybe you might impress was, this shiny man maybe I like it a little bit better. But is that worth hours and days of your life, you're wasting time. I'm trying to get you to not waste time here. Outstanding questions, frequently, a lot of my questions are legal.

[00:07:53]
Like, hey, I don't know if we can secure content rights for this. And then I put a technical one in here is, hey, are we gonna do this over? WebRTC, are we gonna do a vendor thing are we gonna build something in house? We haven't decided that yet that's a big open question, I'll put that in there as well.

[00:08:10]
And then the appendix here I just put a bunch of links to other docs I didn't write these docs, but imagine these docs existed. Here is one that talks about off-the-shelf vendors that we could look at. Here is one that is, what is Hulu, what is HBO Max doing in this space, what is Amazon Prime doing in this space?

[00:08:32]
And here is a retrospective of the Xbox app that we launched once upon a time and the Windows after launch once upon a time and why this is different from them, right? That doesn't need to go into product spec, but I can imagine someone saying, haven't we tried this before?

[00:08:48]
Why is this different? And I was like, let me point you to the doc that says, hey, we've tried this before, why is this different, right? Cool, and that's a pretty basic product spec that I feel pretty okay going into a product review with, hey, here's kind of everything that I'm thinking about here.

[00:09:07]
What do you all think about this? There's plenty of here to discuss and it keeps it pretty focused on. Here's what I wanna build, here's how I wanna build it, here's why I wanna build it. What do you all think about this?