Feb 5, 2015

Taking Assembly for a Test Drive

I have been taking the web product Assembly for a test drive.

Assembly is a platform for crowd-sourced web-based products.  A product for product development online.  Sounds like recursion!

It's got to be a dream for programmers and web geeks!

Product Description and Traction


Assembly is really interesting, just because you get to see what other people are thinking about, in a transparent way.  The product seems to have solid traction based on cosmetic review of the activity.  Here are some key observations about this product.

Coderwall Project bounties and contributors (from Assembly)
  • Virtual - The projects are mostly virtual from around the world.  The landing page is filled with many approved and pending projects.  A large product like Coderwall has 100 contributors, although contribution follows a steep power law distribution (probably even more than 80/20 rule - meaning 5% do 95% of the work ... the nice thing is, they are paid accordingly). 
  • Platform - The diversity of products is as wide as people's imagination.  Of course, what Assembly provides are the building blocks (see below) of a web product building.
  • Revenue Positive - As far as I could tell, Coderwall is its most profitable product; based on 2014 payouts, it seems to have about $24K MRR with 86% profit margin.  Wow!
  • Engineer Heavy - I've joined a few products and started one to test the default experience.  I don't have hard figures, but based on new people who sign-up to a project, the site is engineer heavy, followed by designers, followed by non-technical people.

What Are the Building Blocks of a Web Product?


One of the reasons Assembly interested me was that the product makes you think about what the building blocks of a web product development are.

https://assembly.com/discover
In essence, Team Assembly would have had to deconstruct the core elements of software development, and so doing, virtualized and abstracted a traditional coming together of a team.  Abstraction and virtualization are both long-term trends in the software-centric world.

So, apart from that interesting academic question, here's my hypothesis about what Assembly believes are core building blocks.
  • Software engineers - Which is great, because a platform like this would be self-selecting to software engineers who are used to working "in the web."  Remote working is a norm in this community.
  • Comprehensive communication - Communication is key in product management.  Many good vs. bad outcome can be traced back to good vs. bad communication.  Assembly uses Slack threads for each project you join.  Seems like a good solution.
  • Project management - This seems to be a missing piece of the platform.  Teams likely will use a Trello board or Google docs outside of Assembly.  The communication threads are nice, but not sufficient for serious collaboration. 
  • Version control - Assembly-built products are linked to Github repositories.  There is no separate bug tracking here, but can be managed through communication threads and pull requests.
  • Incentives - I think Assembly's solution is particularly novel and effective.  I think it is very meritocratic, because coins are awarded based on contribution.  The overall model is very Silicon-Valley-esque. The founder holds the lion's share of initial coins, and if everyone maxes out the contribution efforts, the founder's holdings can be diluted down to 50%, and everyone else holding 50% ... unless founder herself is doing a lot of work and awards herself.  Apart from monetary incentive, the product also leverage's people's sense of curiosity and personal value. You can say, "hey, I'm only a designer, but I can add this little piece to the overall effort!"  And so on.  For many, I would say, money is a distant second or third incentive.
  • Team Trust - This is another challenge.  Trust is really hard.  Most startup teams at least, fail because of bad founder selection.  Here, the process is just left up to the wind.
What else is important to building a good product?  Really, it is customers, and a good process to deliver what those customers want.  That's it.  If you don't have customers, things like project management, engineers, and everything else doesn't matter.  Only thing is, some startups start out with a hypothesis they can get customers, and start building.  But, in the end, you need customers.  Communication, version control ... don't get lost in the details.  

It's just tools and processes to serve the customer.

Assembly's mix of light-weight and mostly off the shelf tools is an interesting approach to discovering better ways to build for the customer.

You Might Also Like ...


Are you a product manager?  What's your philosophy and approach to building great products?

You might also like the following teardowns and meta talks on products:

5 comments:

Peter Shin said...

Thanks for the write-up! I'm more intrigued than ever and I will definitely give it a test drive soon.

Unknown said...

Thanks, Peter! Glad you found some use!

Aleem said...
This comment has been removed by the author.
Aleem said...


Hi, I do believe this is a great blog. I stumbledupon it ;) I may return yet again since i have book-marked it. Money and freedom is the greatest way to change, may you be rich and continue to guide other people. Instacart promo

markson said...

Well, there is no platform who answers better than this platform. The question was what is test strategy seems simple but it aint.Here a link to the answer.