Software Engineering · Release Management

What are five things I can expect in the first two weeks of releasing software?

Mike Whitfield Sr. Software Engineer, EPAM, Google

January 3rd, 2015


I've poured my heart and soul into some software, and I developed it 100% by myself so it feels rather scary to just release it to the web-at-large. I'm specifically interested in feedback from folks that have had to release software to the WWW or who have a customer experience department to manage customer feedback and metrics. I have experience with getting the word out, so I'm familiar with traffic patterns but what else might I expect? Maybe most important (and unseen), how am I going to feel throughout this process? I've slaved away on teams before but have mostly been in a support role, so it would be interesting to hear from the executives that took on all the risk before a big release. Surely one can't expect to be doted on by simply releasing the software (c'mon, most people *hate* Mark Zuckerberg). From what consulting experience I have, it seems like the very goal of user adoption is inducing or supporting neediness from the customer. To me, this might make the feelings of releasing softwre more bitter than rewarding. What are some ways you have found to make this process (given my assumptions are correct) not so bleak?

What do you find most rewarding about the software release process? What can you rely on to feel rewarding in a software release and how do you go about seeing the fruits of your labor? I'm guessing it could be things like money, number of downloads, positive reviews, etc.

Feedback on this would greatly help me, thanks so much!

Varun Mehta CEO of Disqovery

January 3rd, 2015

  • For a few days you are going to check the metrics obsessively. I even downloaded mobile apps for MailChimp & Google Analytics during that time.
  • If you do invites over e-mail, you're going to see if people whom you know actually opened your e-mails, and who actually signs up/converts. You might feel slighted when a friend had said, "This sounds great I will definitely sign up" and then he doesn't. You'll grow a thicker skin.
  • You're going to get random people you've never met signing up to try it out. A small percentage of your initial users will be super-engaged. You'll reach out to some of these people and be amazed by how much they love your product and what they're doing with it. This will feel awesome.
There's probably more, but since these are the first points that came to mind (almost 2 years later), they are likely the strongest.


January 3rd, 2015

This post hits too close to home.  

I too have recently ditched the world, went underground from my family and everyone and just coded my heart out.  It was an amazingly rewarding time and the stack is ohhhh so delicious.

I built something I think is AMAZING and changes the world in a small but meaningful way.  I emerged from that place with my very easy, very useful to the right people product and yelled it as loudly to the world as I could!  For over 4 months now I have been showing and revising and showing off the product.  For 4 months it has been at 100% a market ready state having no "technology" challenges keeping it from success, it was sales now.  Well I thought.  But it turns out sales and marketing are not similar things.

Pushing and pushing and pushing to make everyone I could possibly touch know about it.  Spent money on PPC and exposure advertising.  Socialized, got resocialized.  Bought people lunch, met endlessly with folks.  Traveled all over the place showing it off and every single person says, YES THIS IS AMAZING, THIS IS CRAZY AMAZING.  Then.  Not a single sale.

Turns out for some reason everyone over 30 thinks it is so wildly mysterious and magical that they actually want me to come sit at their events and use the product for them.  The room and everyone says yes, yes, wow.  

So...   For me it turns out, there is not 1 single thing I cannot handle from the technology vision, stack and production stand point.  If I had a commercial on HULU tomorrow that sold 10,000 copies that were in immediate use to serve 400,000 people I could make this code stand on its head and sing for the nice audiences.  Heck I can probably even sell ice to Eskimos.   But I clearly cannot "Market" any better than I know the rules of cricket.

So for me, I am looking first thing this year to get back to the grind stone and begin professional working again.  I have this amazing secret product in my back pocket and it has been nothing but frustrating to find a foothold.  I was headed to build 10 more of these amazingly small, low cognitive friction - high value multi-platform products but have stopped with the first one realizing that I DO NOT have what it takes to solve this entire puzzle by myself.  If I keep going I will have 3 products I cannot sell and really be up a creek!

Thanks for the CodingHorror article Mike, Jeff Atwood is a rock star!

Please share your journey as you go, best of luck finding your stride.

Mike Whitfield Sr. Software Engineer, EPAM, Google

January 3rd, 2015

Found this, and it may help:

Samuel Brown Senior Product Architect at InboundCrowd

January 6th, 2015

First off, kudos for taking the big leap to produce your own software and release it to the world.  It is a big step and one that requires a lot of dedication.

A word of caution, though...don't fall in love with your creation.  I know that seems counter-intuitive but you are obviously not the consumer of your software.  Your main (and probably only) focus should be on customer success and happiness.  Red flags go off in my head whenever someone in charge of a product refers to customer 'neediness'.  The fact is, no one 'needs' your product unless you've made it useful and usable enough for them to enjoy it.  

The products I use most (and buy) are very customer focused and excel at constantly refining their product based on feedback.  As I said above, do NOT fall in love with your current product and instead really dive into what your customers want and need. Solicit feedback early and often and make changes that seem sensible.  It may be that the end result of what you create is vastly different than what you envisioned but you have an extremely satisfied (and lucrative) customer base.  This is the heart of 'Lean Startup' and if you haven't read it I highly recommend it.

You've started on a great path and once again, congratulations on doing something many can't or won't embark upon.  Sounds like you've got the passion, best of luck!

Terrance Cohen Senior Software Engineer at Riot Games

January 6th, 2015

A few that come to mind: 
1.Your analytics are missing significant stats and KPIs, so you don't know if users are doing what you expected. 
2. Users aren't doing what you expected. 
3. Some of the systems you thought were critical and/or robust are really trivial and/or fragile, and vice-versa. 
4. Some of the things you thought were really important didn't ship. 
5. That's a good thing, because you shipped.

Owen Rubel Creator of API Chaining, IO State, API I/O Abstraction and modern API Automation

January 7th, 2015

Bingo to all the above. Expect zero to minimal traffic; no one knows your brand. They may have seen your product but it all depends on your brand.

Second, it it depends on how accessible you have made your product. Is there a free version as well as a paid version? Do they have to register and if so, can they register using social media? social media registration with Linkedin, Facebook, Googleplus have a 90% clickthrough rate vs your own registration which acts as a WALL.

Do they have to do further registration or configuration? Are you asking for marketing data or personal information? People may give up at this point and you will lose a approx 15-25% of your clients (or more depending on the area and industry).

You will then spend the next YEAR tweaking your product, your marketing and your branding trying to figure out how to increase your market share.

At which point you will figure out what is wrong with branding, product or marketing... or you will fail.

Candice Hughes, PhD, MBA

January 7th, 2015

It seems to me that the question is wrong or maybe in the wrong order. First, even before building the product, you should have been talking to customers and finding out what they need and want. Customers don't always know exactly, but they can give you a ballpark idea and tell you their desires and strong dislikes. 

Once you have that (plus names of potential customers) then you build, tweak, build, tweak. (Getting customer input along the way)

While you're doing that, you're building your brand, interacting with consumers frequently. This is the process of paving the way, beginning to get brand recognition, finding out the best ways to reach potential customers. 

Then you get to the point where you have a good enough product you can release. You plan the release around a marketing campaign that begins several weeks before release. You build up anticipation, keep building brand and anticipation. 

When the big day comes you shout out to every customer involved all along the way. You ask them to become brand ambassadors and tell their friends. 

You put your heart and soul into branding, branding, branding, in a way that helps your customers. Your product meets some deep need you uncovered earlier. They can't live without it. (If not you made a mistake earlier.) You work day and night getting your product into influencer's hands and in front of customers and keep going until the traction starts tipping your way. You expect it will be really really hard to be recognized and heard in the vast noise of our overinformed world and keep plugging away.

I wish there was a magic bullet. Maybe there is if you're the lucky superconnected person in Silicon Valley. For the rest of us, not so much.

Mike Whitfield Sr. Software Engineer, EPAM, Google

January 9th, 2015

Thanks everyone who responded.

It's impossible to embark on creating complex software without tying your emotions in, else why do it?  There are a couple market strategies, but most of all they focus around being personally sustainable.  I'm a huge advocate of personal development, and I maintain it's paramount for anything in life.

What I have found to fall "out of love" with the product is to simply give myself time.  I get better at watching the market, but as some folks have said you get no where not shipping.  Still, in the specific segment of open source software I'm in, there are people that get squashed all the time for having put years of work in without planning their defenses or revenue stream appropriately.  My customers can become my competition real quick, so in the open source space differentiation and competitive advantage is key.  The point is that by selling a tool, you allow others to build their own constructions with that tool.  Another obvious adage is one my professor gave me, "how many times can you reinvent the toaster?"

In the end of any scientific field, sometimes effective cancer treatments don't ship.  Life can't be more infuriating than for that scientist, so I'm alright.

I think I'd like to see a return to the waterfall style of development, or in computer science terms (and you can quote me on this), we will shift from a stream-based environment to a batch-based environment.  The world will undergo a tremendous amount of social change while this happens, and the shifts will no longer come from technology but the users of the technology.  The point is that customers are unreliable, or the reliable metrics are continually expanding outward away from technology so that it gets harder and harder to align the technology with the user.

The rest of my personal items have left me in a baffling scenario.  I have made significant bets in my career to dump my energies into R&D, and while there is the appearance of similar solutions we all know a true effort can't be created from marketing alone.  I'm detaching my personal feelings from the product, and simultaneously if I don't give it more push nothing at all will result.  Financially, I am asking for the market to take me back in but I swear I'm neither qualified as an engineer or a manager any longer and the resume gaps leave hiring managers with raised eyebrows.  In this way, people might wonder how they can learn from what I've done or take me apart in some respect to scrap my efforts.

In the end, it seems like the entrepreneurial path is to choose one's own demise.  I know release is coming, it's just a matter of time.  Recognizing how unimportant the work I do has been part of the process.  One man's garbage is another man's treasure seems to be the prescient phrase to operate with.