Dogfooding to Build for Creators
Become a real user of your own product so you feel the problems instead of just reading feature requests
- Difficulty
- Easy
- Time to result
- ~months to results
- Steps
- 3
- Confidence
- 95%
When building tools for creators (or any product where the user's livelihood or emotional investment is on the line), you cannot prioritize correctly from data and feature-request lists alone. Maya requires her PMs and engineers to genuinely use the product — for podcasting tools, that means each person starts and runs their own real podcast — until the user's mindset 'clicks' and they can advocate for the right roadmap.
Origin
Maya Prohovnik's own practice, formed from starting in customer support and running four of her own podcasts as employee #1 at Anchor; she cites Airbnb's Brian Chesky pushing the same 'be a host' principle.
Core principles
- 01You can't build the right things if you don't understand the user's mindset from the inside
- 02Feature-request lists get prioritized differently than problems you deeply feel yourself
- 03Building for creators is like B2B: it's someone's business and livelihood
- 04Elevate and celebrate teammates who dogfood so it spreads culturally
How to run it
- 1
Make yourself a real user, not a test user
Actually create and maintain genuine output on your own platform. Maya built real, highly-rated podcasts rather than a throwaway 30-second test clip, so she experienced the true emotional barriers to entry.
Pro tip Pick a topic you could 'talk about forever' so you actually sustain the effort long enough to learn.
Watch out Recording a 30-second test and publishing doesn't count — you won't feel the real pain your users feel.
- 2
Push the whole team to dogfood, not just leadership
Require PMs and engineers to use the product. Incorporate product creation into off-sites and team activities so everyone is forced through the real experience.
Pro tip Listen to and support the content/output your teammates create — it makes dogfooding socially rewarded.
Watch out Teams that only 'talk to users and see the data' will keep missing what those users actually need.
- 3
Convert your lived experience into product and onboarding
Turn the barriers and tricks you discover into a shareable playbook and, ultimately, into onboarding education inside the product (e.g. 'record with a friend,' 'don't follow a rigid script').
Pro tip Your personal onboarding tips are candidate copy for the product's own onboarding flow.
In the wild
Maya's PMs talked to users and saw the data but didn't truly understand needs. Once each of them finally started their own podcast, 'something clicked' and they could build the right things.
→ Team shifted from feature-list prioritization to advocating for real user problems as a big chunk of the roadmap.
Common mistakes
Leaning only on data and feature requests
Without feeling the problems yourself, leadership defaults to 'the next big strategic thing that's two years away' while current users get no value and leave.
Treating dogfooding as a token exercise
A 30-second test clip or a fake account skips the emotional journey (putting your voice out there, wondering if anyone's listening) that is the actual insight.
Is it for you?
Best for
Product leaders building tools for creators, or any B2B/livelihood product where users have deep emotional or financial investment
Not ideal for
Products the team genuinely cannot use authentically, or purely infrastructural back-end work with no meaningful user-facing experience
From the transcript
“all of them once they finally start doing their podcast they're like I get it like something clicked and now I feel like I really…”
“I really don't think that you can build the right things if you're you know they talk to users all the time they see the…”
“becoming one of the users I think it's made it so much easier for me to advocate for that being you know a (13:00) pretty…”
From the episode
Building Anchor, selling to Spotify, and lessons learned
Maya Prohovnik (Spotify’s Head of Podcast Product)