How Ashutosh Priyadarshy designs a focused, bullshit-free workday

Executive overview

Most workdays fill up with activity that is tangentially relevant at best. Ashutosh Priyadarshy, founder of daily planner app Sansama, uses a simple filter: if a task is two or three steps away from a direct result, discard it.

His system combines thematic days, double-buffered time estimates, and intentional planning — not to do more, but to do only what's obviously worth doing.

If you can articulate exactly why something matters today, it belongs in your day. If you can't, it probably doesn't.

Sleep and morning routine

• Sleeps without an alarm; wakes naturally when rested
• Environment stack: blackout blinds, fan for cooling, earplugs for noise
• Slow 20-minute morning boot-up — coffee first, no fixed sequence
• Mood-matched micro-routines rather than a rigid morning script
• Treats the morning as a reaction to the day's conditions, not a fixed playbook

Daily planning ritual

• Plans lightly the night before: lists intended tasks in Sansama
• Reviews and adjusts the next morning once new requests have arrived
• Adds time estimates to every task, including small ones
• Targets a total of ~7 hours of planned work per day — his observed capacity for focused output
• Overestimates task duration deliberately to build in natural buffer

Thematic weekly structure

• Monday: miscellaneous tasks, team meeting, throughput-focused
• Tuesday: entirely customer-facing — calls, support tickets, surveys
• Wednesday–Thursday: deep work only, zero meetings
• Friday: reflection, metrics review, async one-on-ones, co-founder planning
• Meetings and deep work are kept on separate days; only ~2 days per week mix both

The bullshit filter

• Ask of every planned task: is this obviously and directly correlated with the result I want?
• Two or three steps of causal distance = likely bullshit
• Example: a networking event might eventually reach a potential customer — but the chain is too indirect
• Daily planning builds the habit; each day of practice makes the filter more automatic
• Saying yes to a bad meeting is a practice rep for the wrong behaviour — awareness builds over time

Parkinson's law in practice

• Pick intentionally short time windows to force creative compression
• Early Sansama team ran two-day sprints: commit Monday morning, ship Tuesday evening; repeat Wednesday–Thursday
• Short sprints forced the team to decompose hard problems into buildable slices
• Time-boxing a writing task to one hour produces the same output as an open-ended block — without the drift
• Planning a 7-hour day creates a motivational flywheel: a believable target you're racing to hit

Feature prioritisation at a startup

• Generate candidate ideas, immediately discard anything that doesn't feel obvious
• Anchor decisions to one guiding metric at a time (e.g. activation rate — users who plan two days have 50% upgrade rate)
• Evaluate each feature against that single metric; swap the metric when the question changes
• Accept that you can't predict which features move the needle — experiment constantly
• At five people, agility replaces quarterly roadmaps; new customer data invalidates plans within weeks

Handling digital distraction

• Does not block or suppress distractions; accepts them as part of the environment
• Shifts focus to what he wants to do, not away from what he doesn't
• Sharing his daily plan with colleagues creates external commitment — natural motivation to follow through
• Distraction breaks are short because meaningful work is more satisfying than scrolling
• If a distraction feels more compelling than the work, the work itself may be worth reconsidering

Low-motivation days

• Diagnoses the cause before choosing a response (poor sleep, overload, etc.)
• On flat days: start with inbox or simple tasks, take a midday break, then re-engage
• Does not force high-output work onto low-energy states
• Pushes through occasionally, but listens to the body more than the schedule

Productivity as output, not hours

• Measures success by whether he is working on the right things during his best hours
• Does not track hours as a proxy for results
• Accepts that task duration and outcomes are inherently unpredictable — control what you can
• Output quality depends on matching energy peaks to high-priority work