The first time I locked myself at my coworking space to finish DenseWiki, it was a struggle, but I managed to do what I set out to do: produce a finished version of DenseWiki’s browser extension.
But it turns out that although it worked well as a beta product, it still lacks the finesse I want from it.
Let’s craft an incredibly magical digital experience for the end user, that eventually becomes second nature.
Cal Newport’s Deep Work
First, let’s work on some practical tips he suggested in his book (no point reading if you don’t use it). I will create checklists for the following four questions:
- Decide on a wildly important goal
- Act on the lead measures
- Keep a compelling scoreboard
- Maintain a cadence of accountability
- Decide where you’ll do deep work and for how long
- How you’ll work once you start to work (checklist)
- How you’ll support your work
Wildly Important Goal (WIG)
Take DenseWiki to 100,000 scholars by the end of 2025. Doing 100,000 lookups per day.
Evoking so much gratitude and dependency and excitement that they can’t pay us quickly enough.
In fact, the goal is to have a product so good that I get to 1 MILLION scholars by the end of this year. So the product, the marketing, AND the branding should all be so undeniably good that there is a clear shot to that path.
Lead Measures
In order to achieve this result for ME, the product has to deliver a knockout experience for SCHOLARS.
To ensure that, there are two major tracks where activities must be directed:
- Improving the product and its onboarding
- Pushing the product to its limits (understanding difficult papers)
A single lead measure would be to increase the number of deep hours per day (rounded).
If in a given hour of deep work, I end up making a big step (product deployment or finish a paper), then I will add a cross etc within that circle of deep work.
Compelling Scoreboard
A bad score board is:
- Not easily visible all the time
- Not up to date
- Hard to update
- Confusing
- Doesn’t motivate / the score never moves or keeps moving meaninglessly
- Shows only lead measures, but no lag measures
So, my scoreboard:
- Made from paper, and connected to my laptop (whenever I sit to work, it shall be right in front of me). I’ll use my WEEKLY PLANNER sheet for this.
- Easy to add a circle for every new hour of deep work, using a marker / sketch pen. For a half day, it could be a circle that’s filled in.
- At the end of a day, I’ll stretch a line under the circles of that day. Then do it again next day.
- At the end of a week, I can count how many deep hours I got that week, and at the end of a month, how many hours that MONTH!
- The goal should be easy to see, because I can visually estimate how many circles there are in a day or a week! And it should be highly motivating to watch the circles multiply over time, along with the count numbers for each week.
- At the top of the scoreboard, add a simple X for every 100 people who join the platform. After X becomes too simple, I can switch to a differently colored X, but keep adding them!
A cadence of accountability
Here’s what a BAD cadence of accountability would be:
- Irregular
- Unstructured
- Doesn’t result in any action items
- Doesn’t include both lead and lag measures
So here’s what mine would be:
- End of every day (quick tally)
- Structure: count number of deep work hours, compare with recent days. Also count if there’s a bump in the X’s, and add that at the end.
- Action items: what could I do to nudge it up a little?
- End of every week (sunday report)
- Structure: tally hours and compare the whole week’s row; update scoreboard
- Think about what I could do to bring the numbers up
Where and How Long
Let’s start with inversion: where I will avoid working.
- Will NOT work in a moving vehicle
- Will NOT work in my bedroom
- Will NOT work in the common area downstairs
- Will NOT work anywhere I might see beautiful women by lifting my head
This means I won’t do deep work at home, and wherever I will sit down to work, I will face away from other people as much as possible.
How I’ll support my work
Let’s invert: what gets in the way of my work?
- Food / hunger / thirst
- “Oh I was supposed to do that” (open loops)
- Phone calls / messages
- Anxiety
- Google searches about something random that came to my head
- Rabbit hole from a Google search
- Meals pre-decided, prepared ahead of time as much as possible.
- Water prepared.
- Workout schedule pre-decided. No guesswork.
- Decision: No other commitments can come in between.
SAY NO to anyone asking me for my time or energy. - Schedule for open loop tasks that are gnawing at me.
- Starting ritual pre-decided, made into a checklist.
- Choice of music pre-decided, set up on YouTube (brain.fm)
- Any google search ideas that come up, I’ll put them in a sticky note
How I’ll start deep work
What’s a bad way to start?
- Tabs that do NOT relate to deep work are open
- Distracting things on my phone that I haven’t resolved (messages not replied to, etc)
- Phone is not on silent
- Need to decide what I’m doing or not doing
- Have clear lead measures I’ll push in that session
Checklist:
- Close ALL tabs except:
- Replit
- DenseWiki
- Whatever paper I’m reading
- Gemini / ChatGPT (either a clean new chat, or one that’s related to programming)
- Phone on silent
- Check any meetings on the horizon, set alarm
- When is my next meal? Write down time, set alarm.
- When is my next workout? Write down time, set alarm.
- Pray (for achieving my WIG, and many WIGs after that)
- Get programming!
The Sprint
https://workflowy.com/#/2011200eba82
Function (cheese) first, form (design) second
- Smooth onboarding and usage (reduce steps as much as possible)
- Visit website (densewiki.com) and download extension.
- One click login / easy signup
- History of concepts looked up / seen
- Button to add explanation from concept page
- Gamification
- Enable instant badges for dopamine (on a given day, if you do X+ lookups, you get a “hardcore” badge)
- Calendar view where you can see which days you got which badge (or none)
- Beautiful stats
- Concept visualization (how your knowledge is growing)
- Share profile (public view)
- Share concept (public view)
Let’s roll!