The first 90% of a project takes 90% of the time and the last 10% takes the another 200%
February 29, 2008 at 8:05 am | In Uncategorized | 10 CommentsWe have been working on our Social Marketplace “Onista” for quite sometime and every week (and every month) we think we are almost ready for launch “only if we fix these remaining little issues”. To be honest, there is no such thing called “little issues”. There are bugs in product (tons of bugs) and we must fix those before launching.
The major lesson I am learning with this experience is that “The first 90% of a project takes 90% of the time and the last 10% takes the another 200%”. For us the last 10% of the project is taking forever and I really hope we finish in next 6-8 weeks so we can FINALLY launch.
Following are some lessons I learned and I am making serious notes to myself about these lessons so I can try to plan better next time
- Last 10% of the project will suck up maximum energy, patience, time, blood out of you, so be prepared for that. Always think about this fact before promising anyone (especially VC) that you will launch in next 4 weeks
- Projects can be done “Fast”, “Cheap”, and with “Good Quality” but you can have any of these two and not all three. Seriously this is so true. In our case since we did not have lot of money, we wanted “Cheap” and also we wanted “Good Quality” so we CAN NOT have “Fast”. But this is f****** really really slow
- You should “Seriously” limit the features that you want to have in your product. When we were planning for building Onista, we wanted to have all. We wanted to be much better than eBay, Facebook, Craigslist, Amazon and hence we kept on adding nice feature list in our PRD. Now we are paying the price during implementation. It is taking long time because its f****** HUGE project that needs team of at least 50 engineers and we are 3 (that too moonlighting)
- A well-planned project will take twice as long as expected and a badly planned project will take four times longer than expected. We are in later category. Seriously the plan was BAD to have so many feature in first place. What was I thinking when I thought I can build something better than eBay and Facebook in 3 months while moonlighting?
- Elance and RentACoder is crap (except for testing services). Don’t go that way unless you want someone to write a quick php login script. (I am thinking of putting a project there for developing a new Operating System like “Clone of Windows XP” with budget of $500 to see if someone really bids on my project.)
- Sometimes it is a good idea to get rid of the problem instead of solving it. I am thinking about taking new approach now (my partners don’t know yet) to resolve these hard bugs quickly. Instead of fixing the bug get rid of the feature itself. cool right?
- No project has ever finished on time, within budget, to requirement – ours won’t be the first to. So be patient
On serious note, YES We will launch soon. We are confident that we will resolve these last 10% of the issues soon and come out with killer product.
Wait and watch.
10 Comments »
RSS feed for comments on this post. TrackBack URI
Leave a comment
-
Recent Comments
Blog at WordPress.com. | Theme: Pool by Borja Fernandez.
Entries and comments feeds.
[...] Onista wrote an interesting post today on The first 90% of a project takes 90% of the time and the last 10% takes the another 200%Here’s a quick excerptWe wanted to be much better than eBay, Facebook, Craigslist, Amazon and hence we kept on adding nice feature list in our PRD….What was I thinking when I thought I can build something better than eBay and Facebook in 3 months while moonlighting?… [...]
Pingback by Facebook » The first 90% of a project takes 90% of the time and the last 10% takes the another 200% — February 29, 2008 #
You hit a nerve that is probably shared by many trying to launch. I never comment on blogs but this one subject is just too close to home to resist a “me too” comment.
It seems you and I have made nearly all of the same mistakes and are both engaged in the ever shifting launch. Its now a running joke. The “two weeks” and we will be ready is a constant source of amusement for friends and family. Good natured amusement but still gets a little old.
I agree with removing features. Even after trying to slim the product down to its bare features, I am still building a very large beast. I could have started with about half of the app’s functionality but that is now with 20/20 hindsight.
I also have had experience w. RentACoder and the like. I have gone through 2 different dev teams and 3 different UX/UI people. My only take away is that you are gambling every time you hire an unknown vendor and you need to try to cut ties when they are not performing. Easy to say but difficult to do in practice. I am getting slightly better through this experience. I have also learned my best people have come through personal referrals but that is certainly not a revelation.
My last takeaway which you did not mention is early prototyping and testing. I have read many books on the subject ( my favorite: Steve Krug’s ,Don’t Make Me Think!)but did not follow through with actions. My one promise to myself is to do paper prototyping early and then do it again after revisions before I spend any time or money on UI or coding. Foolish to do otherwise.
Anyway, good luck and thanks for the article. Its nice to find company while in the midst of challenges.
Comment by Greg — February 29, 2008 #
If it’s a web app — launch anyways to semi-closed beta.
User feedback is more valuable than bug free code until you have a strong user base, and deploying new updates is simpler with web apps than with desktop apps
Comment by engtech — February 29, 2008 #
Everything in the article is so true. I have been through many projects and I have experienced almost everything mentioned in article.
I suggest you should remove unwanted features and release the product even with 10% bugs. Users will tell you what they want you to fix and what not.
be patient, be focused, and you will succeed
Comment by Peter — February 29, 2008 #
Check out the e-book Getting Real it give some great suggestions on how to structure a development cycle to push thru to deployment.
Comment by paulmobley — February 29, 2008 #
You need agile. Try starting here: http://www.amazon.com/Lean-Software-Development-Agile-Toolkit/dp/0321150783
Comment by Eric — February 29, 2008 #
Two anonymous cents:
You’re in one of two situations. If you’re actually 90% done, meaning basic functionality is solid, then you’re 100% ready to launch. Launch as beta, set a tight secondary deadline, and get as much of the last 10% done as you can. Eventually you will collapse and/or start a new project, so remember to comment your code, even at 4 am.
On the other hand, it might seem like you’re 90% there, since you’ve burned through most of your project’s time/money. However, if a major piece of the puzzle is missing, you’re 50% of the way to launch, due to low team morale + fatigue + financial/managerial pressures. The best option here is to give everyone one “vacation weekend” – no working from home, no work-related email, no staying late Friday, just chill out for two days – then reconvene, set a realistic strategy (not just ‘find bug, fix, launch’), and push very hard.
You’ll deal with both scenarios in life, and both are learning experiences. Enjoy the challenge and hang in there.
Comment by Surfer — February 29, 2008 #
[...] The first 90% of a project takes 90% of the time and the last 10% takes the another 200% We have been working on our Social Marketplace “Onista” for quite sometime and every week (and every month) […] [...]
Pingback by Top Posts « WordPress.com — February 29, 2008 #
thats it, guy
Comment by Lanfordwv — March 25, 2008 #
Hi! I was surfing and found your blog post… nice! I love your blog.
Cheers! Sandra. R.
Comment by sandrar — September 10, 2009 #