Numbers & Nurture

I was chatting with a friend who was surprised at how often we actually make changes to our codebase. He’s a lawyer and has no context. I struggled to explain it and ended up on a concept that I called “Smallest Shippable Value”. People especially non-technical people understand the value received from software works like this.

1. You get an idea or feature request – say “Update the shopping cart experience”
2. Code project
3. Submit code for review and testing
4. Ship Code – begin value benefit

It looks something like this. You get no value until the whole feature is shipped.

The point is to deliver value to the business earlier if possible, so when reviewing features we find which ways we can ship value before needing to have project completion. It can look more like this:

1. You get an idea or feature request – say “Update the shopping cart experience”
2. Break it down
   1. Put our top selling items in a new section called “Top Sellers”
   2. Move the button from the bottom of the page to side scroll with the customer
   3. Change the checkout button to green
3. Code project –
   1. Developer is assigned first task “Put our top selling items in a new section called ‘Top Sellers’”
4. Submit code for review and testing
   1. Developer finishes “Put our top selling items in a new section called ‘Top Sellers’”
5. Back to coding –
   1. Developer moves onto “Move the button from the bottom of the page to side scroll with the customer”
6. Ship Code
   1. Deploy “Put our top selling items in a new section called ‘Top Sellers’”
   2. Value is already delivered!
7. Submit code for review and testing
   1. Developer finishes “Move the button from the bottom of the page to side scroll with the customer”
8. Back to coding –
   1. Developer moves onto “Change the checkout button to green”
9. Ship Code – “Put our top selling items in a new section called ‘Top Sellers’”

This can be visualized a bit easier:

The REAL secret is finding those things that that are the highest percent of the value that are worth shipping. Shipping every letter change is not embodying SSV and would just weigh your processes down.

Pros:

• Smaller/more manageable deploys
• Better time estimations – I generally don’t ask for time estimation over 1 month, but SSV really helps avoid crazy project scope estimations

Cons:

• Can only happen with a robust CI/CD pipeline
• Deploying too small of value can be counterproductive