Back To Blog
Launch Broken But Not Too Broken—Redux
07.29.11

A couple of months ago, I blogged about a phrase that I particularly like using with startups and product, "Launch Broken, But Not Too Broken."

It's a little bit of a subtle phrase and I was reminded of it when I met with a startup who had just launched a new version of their product -- in the spirit of "launch broken, but not too broken" -- and asked me what I thought. My response? "I would not have launched it."

I'm going to break down the key components of the phrase and where I think folks sometimes run into a little trouble.

Launch: shipping products is hard. Focusing on the launch is a big deal. I know many very good PMs who are simply very slow in launching. Usually it's because they have perfectionist tendencies and nothing is ever quite good enough. This is problematic for a couple of reasons. The most obvious one is that things take way longer than they should to get out the door. The second reason is actually a bigger deal. The assumption that more time spent planning and refining is better than launching a slightly inferior product is generally wrong. The way I feel about a product at each stage: conception, design, internal beta, staging server, live -- I feel differently about it each time. And forget about when you actually get real user feedback. You can't get to those stages without actually being at those stages. So having a mentality around launching is critical. When I was a PM, I had a clock in my head. I think the main reason is because if you're the PM of a product -- the #1 question you get is, "When are you launching?" PMs should feel this pressure -- there should be no satisfaction in getting only intermediate work is done. The goal is more tangible than that -- it's getting a good product out the door.

Broken: To overcome perfectionist tendencies (or fear of failure) -- I think it's useful to use the concept of "broken". It removes the element of failure because the assumption is that what's going to be launched will be screwed up. Then you'll go and fix it and improve it and iterate on it. No big deal. It's just work. It's not a failure and it's not a judgment on your character or ability to perform. It's not only mentally freeing, but more importantly, gets products to more advanced stages more quickly.

But Not Too Broken: Here's where judgment comes in. A PM needs to know what's acceptable and what's not. Judgment is intuition which is built up through self-reflection and experience (which involves a lot of screwing up unfortunately.)

So what's too broken? Here are 3 signs a product is "too broken" in my view:

1. Bad QA
At amazon.com, our concept of QA was this. When the product was about to launch, an email went out and everyone went and clicked on the links on the site. We found bugs and submitted them. (I'm sure there are QA professionals out there that are shuddering reading this.) I literally (even now) will click on every link of a site launch. Every link. It is unbelievable the level of bugs I'll find using this simplistic method. I sometimes do this with major launches of large tech companies and even then find quite a number of bugs. This is boring work but it has to be done. I routinely do this and find: code on the page, broken links, iPhone/Android apps that frequently crash, and so on. This is too broken.

2. Too Difficult to Use
My mom (who is not particularly technologically adept) and my 2-year old nephew both use an iPad. Think about that. Here's a rather complex product that both of them -- spanning age ranges, cognitive ability, and familiarity with technology -- and they're both able to use the iPad with little to no problems. Granted, Apple is probably the gold standard in terms of usability -- but I think that's the standard that all products should strive for. Is it simple enough, clear enough, that someone like my Mom could use this product? I think a big part of being a product manager is being able to project yourself into another person's shoes. How do they think? What are their wants / needs? What's their day like? What would make things easier? Think about your Mom -- is your product clear enough and easy enough to use? Are you using the minimum number of screens to get something accomplished? Are buttons in the most clear place? As an example, think about when you come across a random e-commerce site. The buy button might be in a weird place. The search function doesn't work so well. The browse links are all over the place. Products are not cleanly grouped together. Then think about when you use amazon.com. It's night and day right? All those subtleties mean something and are important. Be maniacal about those details.

3. Value Not Clear
Too often I'll talk to an entrepreneur and they'll have a compelling idea and a real passion for the space. Then I use their product and that idea / passion has not translated into the product. Why? Product development is hard. It's big strategic vision combined with lots of nitty gritty details. The customer has to be able to come to your homepage and immediately understand, "Oh, I understand how that works and why this is valuable for me." I was talking with someone the other day about Groupon's homepage for merchants [link]. Everything on this page is about getting new customers. "Like new customers?" "Learn how a one-day feature on Groupon can bring your business thousands of new customers." There's more details and other links for people to peruse and learn more -- but the core of it is clear. Groupon gets your business new customers. I'm blending both product and marketing here -- but I think ultimately it's the same. What am I looking at and why is this valuable to me?

I frequently remind myself of this phrase. Why is something taking too long? Am I planning too much? Do we have too big a team? Are there others who do equivalent work more quickly that I can learn from? After I complete a project (or even an intermediate step) -- I'll go back and review.

The last thing I would say is that sometimes we don't necessarily know what's broken versus too broken. So get help. When I produced my web series, I got a ton of help. I had half a dozen people give notes on the writing (ranging from professional screenwriters to directors). I sat down with others who have produced before. I got people who do this stuff all the time to work on my project. The end product was significantly better because they were part of the team. Treat product the same way. Identify areas that you're worried about and find people (formally or informally) to lend a hand. Not sure if something is too broken? Find someone you respect on product and ask them for their opinion. Not sure who you should ask? Ask several people -- and then see where that cluster of feedback leads you. Over time, you not only build up a nice network of people you can go to but also your own instincts will get significantly better.

Comments