Prototyping · Minimum Viable Product

Experience using ODesk to build a prototype?

Nathaniel Dean SaaS Sales Executive

August 14th, 2013

Hi FD, does anyone have experience using ODesk to build a prototype product?  I'm looking for positive/negative experiences, non-obvious pitfalls, tips, etc.  

Especially helpful is if anyone has experience managing multiple contractors who may have to work together (e.g. engineer + designer + content creator).  

Thanks!  

Geoff Whitlock Co-Founder and President of Surround

August 14th, 2013

It works great if you are very prepared. Any unanswered questions or ambiguity, and it's a waste.

Greg Belvedere

August 14th, 2013

I used ODesk and I found a great developer to build my MVP. I had tried elance, but found that all the developers that contacted me were from India or Eastern Europe. The messages I received in broken English made me not go that route, because I value clear communication. With ODesk I found several developers I liked, some of them were even in the US. Though I ultimately hired a great developer from the UK. http://thebookelf.com/

John Pearce Information Entrepreneur

August 14th, 2013

Yes, I've had some recent experience with oDesk. It's tricky, promising, full of potential and pitfalls. I think asking people from different cultures and time zones and languages to work together is tough. Communication is a big issue with many of these freelancers. Better may be if you have an extremely clear idea of what you want from each, and manage each one directly, vs. expecting them to work together. Another possibly better alternative would be to hire an oDesk agency that integrates all you need in a team that has worked together before. Then you'd have one point of contact. My last suggestion would be to post the project as specifically as humanly possible, probably seeking an agency as noted. I tried picking people who seemed qualified and inquiring with them directly vs. posting to the oDesk hordes, and on balance I preferred wading through a few dozen applicants to find a strong match. The latter model worked better. We had good luck with a Brazillian developer that way. Ultimately, it's not about oDesk, it's about the person you happen to find via oDesk. They vary wildly, and, as noted, communication is key.

Anonymous

August 15th, 2013

I use the opposite approach - instead of being ruthless or pushing hard, I work hard to 'clear a path' for the developers to succeed, and try to get them into a feeling of partnership rather then a 'you work for me' type relationship. Software is hard enough when you are in the same location - when you go offshore you really need to ensure that the developers are on your side by managing them effectively, not just aggressively.

Wireframing first is definitely a must, though!

Joel Magalnick Storyteller. Innovator. Leader.

August 14th, 2013

I'll second what Geoff said. If you map out everything exactly as you want it, and expect pushback if there's any change, then it can be successful. I've had one good experience and one lousy experience. But you have to be completely clear on your project goals and make sure the developers understand exactly what you want, especially if there's a language barrier.

Benjamin Forgan Founder & CEO at Konekt

August 15th, 2013

use something like teamtreehouse.com to a least understand what language you are building your product in and why. this will also give a decent idea of scope and how much time it should take for someone decently competent to complete. if going hourly, be a ruthless clock manager (4hour work week has some good tips for this), if you agree to something at a fixed price then make sure you push for deadlines and are 110% clear in your directions. Use something like balsamiq to mockup your idea. hope this helps.

Caen Contee Connect.com: Early Stage Startup Streamlining How We Get Together with Friends. Seeking New Investment.

August 14th, 2013

Nathaniel,

I'm not sure what your product is, but I've used an american web development and design agency to great effect.

They have deep experience, capabilities and can offer custom needs depending on the SOW (ie. only development, only backend, all development and design). If you can find the right team this works better than ODesk, which I've also used as they have a clear work flow and in-house project manager, both large pitfals when working with independent contractors.

Happy to make an intro. You can reach me off FD at caen.contee@gmail.com.


Honed daily,
Caen

Anton

August 26th, 2013

@Guy: yes, most developers and even agencies reuse at least some of their code from project to project. It depends, though, on the project. If you get a "good deal" on a fixed-price project to develop a website or app, there's a good chance it's cheap because the developer is repurposing old code with a new skin. Very custom and high end projects have this happen less frequently, of course, but we are talking oDesk here. 

t's not always a problem, as some developers have libraries of code they reuse between projects that lets then work faster and more efficiently. As long as you're getting the design and functionality you need and you contractually own it, I suppose it can be okay. But do check your agreement with whomever you hire. The biggest pitfall with reused code in my experience is that certain features are difficult or costly to add down the line, caused by shortcomings with reused code that was intended for something completely different.

Alexander Ross Head of Business Development at Verifide

August 15th, 2013

Hello Nathaniel- I have done this several times and think it is def a viable approach, with a couple pitfalls of course. I dont have time to write up my thoughts here now but feel free to ping me at alex at presentista dot com. Cheers, Alex

Guy Marion Chief Marketing & Revenue Officer at Autopilot HQ

August 25th, 2013

Question for all: has anyone ever had a remote contractor borrow the code they build for you, for their own purposes (or other clients)?  

My feedback: I've hired about 6 or 7 different oDesk developers over the past 5 years (web & backend developers for Drupal, Python, and a bit of Ruby). I second many of the comments above: about half my projects were successful, but required significant management time (preparedness, clear requirements, defining testing/deployment workflow, building rapport, mutually agreeing on success criteria). The upside is that it is possible to build a website or MVP in 2-3 months affordably, while you're still a lone entrepreneur or busy team. Downsides are that you're working in a more waterfall manner (vs. doing in-house agile development), which can be challenging in your early days. Also, quality of code can be an issue.