Pros and Cons Example: 9 Pros, 5 Cons, and the One That Decided It

Almost every pros and cons tutorial demonstrates the technique on a decision nobody made, with items nobody wrote, and then declares the longer column the winner.

Real lists are rarely published. Lists with a checkable outcome attached are rarer still. This one is both.

In November 2019, six months after leaving the CTO role at Artsy for a Principal Engineer role at AWS, Daniel Doubrovkine published the actual list under his own name: nine pros, five cons, for going from management back to individual contributor.

Seven years later we can ask the only question that really matters about a decision framework: did the structure of the list do any of the work?

It did not. And the reason is a property of the decision that he did write down — in the wrong place.

The list, as published

Pros (9)	Cons (5)
Extended technical relevance by ~a decade, at unprecedented operational scale	Loss of organisational power — no longer in hiring, firing, promoting
Autonomy — "working autonomously is literally in my job description"	Can only influence, not decide, on non-technical matters
Reduced stress; normal hours; room for deep work	Must "spend a lot of energy explaining a problem to a decision maker instead of just fixing it"
Coding as a primary activity	Many perceive the move as career limiting
"I can choose not to take a meeting, cater to an investor or deal with an unhappy customer"	No longer the face of the organisation
Not on call for catastrophic failures
No direct reports venting; less gossip and politics
Competitive compensation — AWS RSUs "are liquid and are the same as cash"
Reversible — can return to management or a CTO role later

Count them and the list says move, nine to five. He moved, and six months in his verdict was "Six months in, yes."

So the framework was right. This is the case that should make pros/cons look good — which is exactly why it is worth taking apart.

What actually decided it

Doubrovkine tells us, and it is not the count:

"My health is better and the stress level is lower. Some things money can't buy."

That is one bullet out of nine. In the list it sits between "coding as a primary activity" and "I can choose not to take a meeting" — three entries that the format renders identical in weight. The thing that decided the decision is typographically indistinguishable from meeting flexibility.

This is the failure mode the framework page already names first: false symmetry, where every entry is treated as equally weighted so the longer column wins. Here the longer column was the right answer, which makes the symmetry problem harder to see, not easier.

A framework that returns the correct answer while contributing nothing to it is more dangerous than one that is obviously wrong — because you will trust it next time, on a decision where the count and the truth point in different directions.

What happened to the cons

Seven years is long enough to grade them.

Con, as listed in 2019	What followed
Loss of organisational power	Real, and by his account it was the trade he wanted
Can only influence, not decide	Real
Energy spent explaining rather than fixing	Real
"Many perceive the move as career limiting"	Did not bind. He became a founding Principal Engineer of OpenSearch, one of AWS's most visible open-source projects, and later moved to Shopify Engineering where, by his own account, he created the Augmented Engineering team
No longer the face of the organisation	Inverted, arguably — OpenSearch made him a public face of a project rather than a company

The con that sounds most frightening when you write it is the one that did not happen. That is not a general law — it is one person, in one market, with an unusually strong track record. But it is a useful corrective to how that particular fear reads on a list, where "career limiting" occupies exactly as much space as "less gossip."

The Decision Replay

What the framework says	What he actually did	What actually happened
Move. Nine pros against five cons is a decisive margin by an unweighted count — the standard way the technique is taught.	Left the CTO role for Principal Engineer at AWS, and published the list six months in.	Stayed on the IC track for years as founding Principal Engineer of OpenSearch; later created a team at Shopify. His own self-description is still "entrepreneur, technologist, and open-source enthusiast." The decisive factor he names — health and stress — held. The scariest con did not materialise.

Right answer. Right outcome. And the list's structure contributed nothing — because the one item that mattered was weighted the same as the eight around it, and the item that should have governed the whole exercise was filed as pro number nine.

The Reversibility Multiplier

Look again at that ninth pro:

Reversible — can return to management or a CTO role later.

He did not omit reversibility. The framework page's third pitfall warns against exactly that — "leaving out reversibility. A decision you can undo in three months should be weighted differently from one you're locked into for five years." He included it.

He included it as a line item, which is the subtler error, because a pros/cons list has no slot for a thing of this type. Reversibility is not a benefit of one option. It does not belong on a side.

The Reversibility Multiplier — reversibility is not an entry in a pros and cons list, it is a scalar on how lopsided the list has to be before you act. Write it above the list as one line: how long would undoing this take, and what would it cost? If the answer is under a year at modest cost, you do not need a decisive list — you need a non-catastrophic one. If the decision is irreversible, a modest majority of pros is nowhere near enough.

Run Doubrovkine's decision that way and the nine-to-five count stops mattering entirely:

1. Above the list: Can I undo this? Yes — a former CTO with a strong track record can return to management. Cost: some time, some perception risk.
2. Therefore the bar is low. The list does not need to be decisive. It needs to contain no catastrophic con.
3. Scan the cons for catastrophe. Loss of power, less authority, more explaining, perception risk, less visibility. All are real, none is unrecoverable.
4. Done. The decision is made without counting anything.

That is a two-minute exercise that reaches the same place as the fourteen-item list, and it reaches it for the right reason. It also explains why he could act on a list whose decisive item he could only articulate as "some things money can't buy" — with the option to return, he did not need to justify the feeling. He only needed the downside to be survivable.

Where this misleads

Reversibility is often overestimated, and the multiplier cuts both ways. "I can always go back" is one of the most common self-deceptions in career decisions — skills decay, networks cool, titles get harder to re-earn with each year away, and the market you would return to is not the one you left. The multiplier is only as good as an honest answer to "how long, and at what cost." If you cannot state both, treat the decision as irreversible.

Doubrovkine's reversibility was unusually real. A former CTO with a public track record and a decade of visible open-source work has options most people do not. Someone stepping off a management track without that record may find the return path genuinely closed, which flips the bar from low to high.

This is a survivorship case with a happy ending. The people whose IC move did turn out to be career limiting are not publishing seven-year retrospectives. The con was real; it simply did not bind here.

One published list is not evidence about lists in general. What it demonstrates is a mechanism — that an unweighted format hides both the decisive item and the governing property — not how often that mechanism changes an outcome.

Now run yours

1. Write the reversibility line first, above everything. How long to undo, at what cost. Be specific and pessimistic.
2. Set the bar from it. Reversible and cheap → the list only has to be non-catastrophic. Irreversible → you need a wide margin and you should probably not be using a pros/cons list at all.
3. Then write the list, and mark the one item you would not trade away. If you cannot name it, you are not ready to decide. If you can, check whether the count agrees — and if it disagrees, believe the item, not the count.

The free Pros and Cons worksheet runs in the browser with no signup, and supports weighting — which is the whole point.

Want to run this on your phone? Framework for iPhone & iPad ships a weighted pros/cons worksheet alongside 100+ other frameworks.

Want to go deeper

• Pros and Cons framework — the method, the four pitfalls, and when to escalate to a weighted matrix
• Weighted Decision Matrix — the right tool once you have more than two options or more than one criterion that matters
• Regret Minimization — the framework for the item you would not trade away
• Ikigai Career Example: The Four Circles Pointed at Google — a Decision Replay where the count and the truth did point in different directions
• Pre-mortem Example: The Failure a Premortem Cannot Catch — the same treatment applied to a startup shutdown
• Ansoff Matrix Example: The Growth Move That Isn't on the Grid — the first business-decision replay, where the winning move is not on the grid

Cover photo: Jens Lelie on Unsplash.