← All Posts

Four Years of Past Year Reviews: What I Actually Learned

Tim Ferriss has publicly stated that the Past Year Review replaced his New Year’s resolutions entirely, and the process takes 30-60 minutes compared to hours of goal-setting that rarely survives February.1

I’ve done the PYR for four consecutive years. The first year surprised me. The second confirmed patterns. The third showed compounding. The fourth changed how I make career decisions.

TL;DR

The Past Year Review examines each week of the past year through a simple lens: what generated positive energy, what generated negative energy. After four years of data, I can report: the process works, but not for the reasons Ferriss emphasizes. The real value isn’t the “do more/do less” lists. It’s the multi-year pattern recognition that reveals which decisions compound and which activities generate diminishing returns. My PYR data directly informed my decision to leave ZipRecruiter after 12 years and transition to independent work.


The Process (30 Minutes, Once Per Year)

Step 1: Scan Your Calendar

Open the calendar from the past 12 months. Week by week, mark items as positive (+), negative (-), or neutral. Don’t overthink it. First reaction only.2

Step 2: Identify Patterns

Group the positives and negatives: - People: Who consistently appeared in positive weeks versus negative ones? - Activities: Which recurring activities generated energy versus drained it? - Projects: Which work produced satisfaction versus frustration? - Environments: Where did the best and worst weeks happen?

Step 3: Create Two Lists

Do more: 5-10 specific activities, people, or environments that consistently appeared in positive weeks.

Do less: 5-10 specific activities, people, or environments that consistently appeared in negative weeks.

Step 4: Schedule the Positives

Book the trips. Schedule the dinners. Block the creative time. Vague intentions don’t survive January; calendar entries persist.3


What Four Years of Data Showed Me

Year 1: The Surprise

My first PYR revealed that my highest-energy weeks weren’t the weeks with the biggest professional wins. They were weeks where I built something with my hands: a weekend project, a prototype, a design exploration. The VP meetings, all-hands presentations, and strategic planning sessions I assumed were energizing scored neutral or negative.

The “do more” list: personal building projects, 1:1 mentoring sessions, long walks. The “do less” list: large group meetings, slide deck preparation, cross-team coordination meetings.

Year 2: The Confirmation

The same patterns held with sharper contrast. Building scored positive 48 out of 52 weeks. Coordinating scored negative 35 out of 52 weeks. The data was clear: I drew energy from creation and spent it on coordination.

A new pattern emerged: learning a new skill (a new programming language, a design tool, a domain) scored positive every single time, regardless of the outcome. The novelty of learning itself generated energy independent of whether the skill proved useful.

Year 3: The Compounding

Items from my year-1 “do more” list that I actually scheduled had compounded. The personal building projects I’d committed to led to three functioning apps. The 1:1 mentoring sessions had deepened into genuine friendships. The long walks had become my primary thinking tool.

Items from my “do less” list that I’d failed to eliminate had also compounded: in the wrong direction. The coordination meetings I couldn’t avoid had expanded. The slide deck preparation had gotten more elaborate, not less.

The lesson: “do more” items compound when scheduled. “Do less” items compound when ignored.

Year 4: The Career Decision

By year four, the pattern was undeniable. My energy came from building, learning, and creating. My energy drained from coordinating, presenting, and managing process. After 12 years as VP of Product Design at ZipRecruiter, I was spending 80% of my time on activities that scored negative and 20% on activities that scored positive.

The PYR didn’t tell me to leave. It showed me the data that made the decision obvious. I transitioned to independent work building 12+ projects across iOS, web, and AI, a schedule where building and learning occupy 90% of my time.4


Why PYR Works Better Than Goal-Setting

Traditional goal-setting imagines a future self. The PYR examines an actual past self. The data comes from lived experience rather than aspirational projection.5

Goals suffer from the “planning fallacy”: humans systematically overestimate their future capacity and underestimate obstacles. The PYR avoids this because the evaluation looks backward at events that already happened.

After four years, I also notice that PYR lists are more honest than goals. A goal like “exercise more” reflects what I think I should want. A PYR positive like “weeks I surfed scored +3 energy” reflects what actually generates energy. The PYR doesn’t care about should. It only tracks did.


A Template That Works

I use a quarterly table in markdown:

| Week   | Events/Activities          | +/- | Notes                |
|--------|---------------------------|-----|----------------------|
| Jan 1  | Team offsite, prototyping | +   | The prototype, not   |
|        |                           |     | the offsite          |
| Jan 8  | Quarterly planning, demos | -   | Too many meetings    |
| Jan 15 | Built search feature      | +   | Flow state all week  |

After completing all 52 weeks, items that appear three or more times across the year are your strongest signals. Items that appear once are noise.


Multi-Year Patterns Worth Tracking

The single-year PYR is useful. The multi-year PYR is transformative.6

What stabilizes: By year two, the core “do more” list stops changing. My core positive activities (building, learning, mentoring) haven’t moved in four years. If an activity appears on the “do more” list for two consecutive years, it’s a durable energy source, not a phase.

What compounds: Relationships deepened through the “do more” list generate new opportunities. The mentoring relationships from year 1 produced collaboration opportunities by year 3. The building projects from year 2 became portfolio pieces by year 4. This mirrors how knowledge compounds across domains.

What reveals itself: Some activities that scored positive in year 1 shifted to neutral by year 3. The novelty wore off, revealing that the energy came from newness, not from the activity itself. The PYR catches this drift before you over-commit.


Key Takeaways

For professionals: - Run the PYR in the last week of December or first week of January; the process requires a complete calendar year of data - Focus on energy patterns rather than achievements; high-energy activities tend to produce better outcomes regardless of initial impressions - Track multi-year trends; single-year data surprises, multi-year data transforms - Schedule “do more” items immediately; unscheduled intentions evaporate by February

For anyone considering a career change: - Run the PYR for two consecutive years before making a major career decision; one year of data is suggestive, two years is conclusive - Pay attention to the ratio of positive-to-negative activities in your current role; if 80% of your time scores negative, no amount of optimization fixes the problem


References


  1. Ferriss, Tim, “Past Year Review,” The Tim Ferriss Show, December 2021. 

  2. Ferriss, Tim, Tools of Titans, Houghton Mifflin Harcourt, 2016. 

  3. Clear, James, Atomic Habits, Avery, 2018. Research on implementation intentions and scheduling as behavior drivers. 

  4. Author’s four consecutive PYR analyses (2022-2025). Energy pattern data directly informed career transition from VP Product Design to independent practice. 

  5. Kahneman, Daniel & Tversky, Amos, “Intuitive Prediction: Biases and Corrective Procedures,” TIMS Studies in Management Science, 12, 313-327, 1979. 

  6. Author’s multi-year PYR data. Patterns stabilized by year 2; compounding effects visible by year 3; career decision informed by year 4. 

Related Posts

Mental Compound Interest: How Cross-Domain Knowledge Built My Best Tools

Design + engineering + AI produced tools none of those skills alone could build. An interactive calculator and the cross…

6 min read

The 15x Cost of a Bad Database Decision: Lessons in Decision Timing

I measured the cost of a database decision across three systems. The migration cost grew 15x over three years. A framewo…

6 min read

The OODA Loop for Prompt Engineering: What Five Failures Taught Me

Five prompt failures taught me that structured observation beats clever wording. Boyd's OODA loop maps directly to the p…

9 min read