The document discusses the evolution of product building over the past decade, focusing on the role of feature flagging tools in managing product success. It outlines who uses feature flags, their applications, types, and important considerations for product managers when selecting a feature flagging solution. A demonstration of the VWO Fullstack tool is also included as part of the content.

1. Feature Flagging
Your Way to
Product Success

2. What are we
covering today?
● How product building has changed in the
last decade
● Intro to feature flags - Who, What, When,
Where
● Considerations for choosing a Feature
Flagging tool - JBTD for product managers
● Demo of VWO FullStack
2

3. How we build product
has changed
dramatically?
1
3

4. Some philosophy first:
4
What is product
success and why it
so matters(a lot)?

5. ● The API Economy - Execution
de-risked
● The rise of Social media and
virality
● Talent density 100X
● End of the beginning - Utility
phase of the internet
5
2021
The cost of
being late
The cost of
being wrong
2010
The cost of
being wrong
The cost of
being late

6. Feature Flags
2.
6
Who? What? Where? When?

7. Origins of feature flagging
● Championed by Martin Fowler
● Also called Feature toggles or Functionality
switches or feature Flippers
● “Conditional code logic” during runtime
without deploying code
● Conditions can be target audience,
timeframe, business metric etc
7
Who?
and
what?

8. Use cases of Feature flags
8
Where?
● Canary releases
● Quicker releases or Testing in PROD or
Kill Switch/Roll back
● Experimentation (AB Testing)
● Feature gating/Operational use cases

9. 9
When?
● Release toggles
● OPS toggles
● Permission toggles
● Experimentation toggles
Types of feature flags: Categorized broadly into
4 types based on longevity and dynamism

10. | 10
PM jobs to be
done with feature
flags
3.
10

11. Product manager
When we release software I want to control the release process so that I
can choose the timing, audience of the release and reduce release risk
● Decouple deployment and release
● Rollback without deployment
● Phased rollout
When we experiment with features I want to have the ability to tweak
experiments without assistance from the dev team so that I can use
real-time data to improve experiments
When we release software I want to control operation toggles without
assistance from the dev team so that I can work efficiently and
independently

12. | 12
4.
12
Buying
considerations

13. Evaluation
criteria
13
Home grown Vs Open source Vs SAAS
● Target using 1st and 3rd party data
● Negligible impact on system performance
● Easy to integrate/maintain for developers
● Fits in your workflow

14. 14
VWO
Fullstack
demo
5.

15. ?
Questions