Skip to content

Commit bedacdc

Browse files
committed
the feedback protocol
1 parent e0deea8 commit bedacdc

8 files changed

Lines changed: 260 additions & 6 deletions

logseq/graphs-txid.edn

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1 @@
1-
["3cc76cb0-46e2-4dd1-8c8a-3ff85e6a2209" "c1c37f1b-4812-416b-bc1b-3cb279754707" 95]
1+
["3cc76cb0-46e2-4dd1-8c8a-3ff85e6a2209" "c1c37f1b-4812-416b-bc1b-3cb279754707" 394]

pages/Cursed frame.md

Lines changed: 4 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -3,10 +3,10 @@
33
- Examples:
44
- Both participants see themselves as superior and the other inferior
55
- [[The depth of any transaction is limited by the depth of the shallower party]]
6-
- Student cannot trust the teacher or no obvious credential of authority to anchor the frame
7-
- Teacher attempting to teach an unwilling student
8-
- Student cannot trust the teacher has their interests at heart
6+
- Example: a failed teacher/student frame
7+
- Teacher attempting to teach an unwilling student; Student does not trust that the teacher has their best interests at heart
8+
- root cause: no dominant credential for student/teacher to coordinate/agree upon which role they play, relationship may not be consensual
99
- A participant doesn't realize what they really want from the exchange
10-
- One person states they want advice but actually wants empathy; other person fails to recognize this and gives rational advice instead of listening.
10+
- "mansplaining": woman states she wants advice but actually wants empathy; man fails to recognize this and gives rational advice instead of listening.
1111
- See: [[Strategy, not self-expression]]
1212
-
Lines changed: 45 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,45 @@
1+
- Powerful phrases
2+
- **You’re right** — Use this even when they’re wrong. This phrase prompts you to remember to find the root of their concern, (which is likely shared by others,) and acknowledge their perspective, which makes them feel heard and therefore willing to now listen.
3+
- [https://www.reddit.com/r/Clojure/comments/11146ic/hyperfiddleelectric_formerly_photon_has_been_made/j8e94pl/](https://www.reddit.com/r/Clojure/comments/11146ic/hyperfiddleelectric_formerly_photon_has_been_made/j8e94pl/)
4+
-
5+
- [https://news.ycombinator.com/item?id=34772833](https://news.ycombinator.com/item?id=34772833)
6+
- [https://news.ycombinator.com/item?id=34776973](https://news.ycombinator.com/item?id=34776973)
7+
- [https://news.ycombinator.com/item?id=31224044](https://news.ycombinator.com/item?id=31224044)
8+
- **Thank you for the feedback** — this is for you as much as them; it prompts you into the right frame of mind.
9+
- **We don’t know yet** —
10+
- **Thank you for the question** —
11+
- [https://www.reddit.com/r/Clojure/comments/vizdcc/hyperfiddlephoton_progress_update/idk0dy5/?context=10000](https://www.reddit.com/r/Clojure/comments/vizdcc/hyperfiddlephoton_progress_update/idk0dy5/?context=10000)
12+
- **“Speak from the I”** – makes it safe to have different opinions, and safe to disagree
13+
- **Never disagree** — see “you’re right”; disagreement puts people into a defensive state.
14+
- words like NO and BUT trigger stress hormone release
15+
- Hey, I like what you did here BUT ...
16+
- **Short replies** — the more you write, the more work you are asking them to do to understand
17+
- **“sarcastic like” “exit like”** – drop a like on an response to make a person go away, you can think of this in your mind as “fuck you :)”
18+
- **Refuting the Central Point** — [PG: How to Disagree](http://www.paulgraham.com/disagree.html).
19+
- you don’t need to quote reply and address every little thing.
20+
- Find the big thing and address it in 3 sentences.
21+
- **Answer with numbers**
22+
- “Engineering is done with numbers. Analysis without numbers is just an opinion”
23+
- [Amazon writing guidelines](https://www.firstprinciples.ventures/insights/four-tips-to-write-like-amazon)
24+
- **Don’t fan the flames**
25+
- you don’t have to answer every question
26+
- you don’t have to answer any question at all
27+
- **Walk away**
28+
- it’s the reply notification that tilts people, especially when the reply is disagreeable
29+
- **let them have the last word** — the real audience is the readers, and they will judge for themselves; you don’t need to “win” or “beat” everyone!
30+
- **Slow down your replies**
31+
- **Don’t get trapped**
32+
- **Be supportive**
33+
- **Refocus the conversation on your talking points**
34+
- **Know the medium:**
35+
- **Clojureverse** — particularly prone to arguments because it encourages long disagreeable replies and emails them, making it hard to ignore. The linear structure also focuses attention on back-and-forth arguments, and interesting little comments get lost.
36+
- **Reddit** — tree structure makes branching and backtracking easy. Small reply box helps make replies shorter.
37+
- **Hacker News** —
38+
- **Twitter** — character limit helps focus. it’s not rude to ignore someone; they will forget. Good mute controls – mute bad threads and bad actors.
39+
- **Kings do not argue with peasants; let masses argue on your behalf:**
40+
- example: [https://news.ycombinator.com/item?id=28630209](https://news.ycombinator.com/item?id=28630209)
41+
- I did not need to say anything in this thread!
42+
- **Strategy, not Self-Expression** — ask, what outcome am I trying to produce?
43+
- or are you just venting your feelings? Get a therapist
44+
- **Say the same thing twice and then give an example**
45+
- **“you count”** — don’t make it about the person, make it about the technology
Lines changed: 25 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,25 @@
1+
- a list of frames and frame collisions
2+
-
3+
- List of communication frames we encounter at work
4+
- teacher/student
5+
- master/apprentice
6+
- expert/expert (collaboration of mutual respect)
7+
- curse of development → clash
8+
- credentialist
9+
- entrepreneur/investor
10+
- boss/employee
11+
- dominant/subordinate
12+
- feedback request
13+
- code review
14+
-
15+
- List of frame collisions
16+
- ...
17+
-
18+
- Hypothesis:
19+
- If a frame is not aligned, no communication can occur
20+
- the student must be willing
21+
- “stalled frame”
22+
- “cursed frame” “frame becomes cursed”
23+
-
24+
- See also
25+
- [[Cursed frame]]

pages/I'm Dustin Getz, digital citizen.md

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -45,6 +45,7 @@
4545
- ## How to start a startup (and succeed)
4646
- [[Aphorisms for Product Management]]
4747
- [[Model: Previously Unknown Axiom]]
48+
- [[The Feedback Protocol]]
4849
- [[Model: Cynefin, a map of uncertainty]]
4950
- [[The CAC Problem]]
5051
- [[DevOps Maturity Levels]]
@@ -60,4 +61,4 @@
6061
- [[On Big Startup Ideas (2017)]]
6162
- [[Coordinating solutions to inadequate eqilibriums]]
6263
- [[N structural issues in venture capital]]
63-
-
64+
- [[Powerful Phrases]]

pages/Powerful Phrases.md

Lines changed: 44 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,44 @@
1+
- A powerful phrase is a way to actively redirect a risky or failing interaction back into a healthy one. It is "powerful" because it **immediately halts a failing encounter before it can do more damage, and transforms it into a positive and productive one.**
2+
-
3+
- Concept taken from the book [Powerful Phrases for Dealing with Difficult People](https://www.amazon.com/Powerful-Phrases-Dealing-Difficult-People-ebook/dp/B00E1SSIV0/), a must read for product managers
4+
-
5+
-
6+
- You’re right “magic compliance extraction” ; “Indeed”
7+
-
8+
- Technical discussions
9+
- I don’t understand the idea → I would like to see a unit test
10+
- It’s been 2 hours → How much more time should we invest today? (Set a timer)
11+
- Frustrated → Should we take a break?
12+
- Loss of composure → I'm not thinking clearly; can I have a moment please?
13+
- taking a break is far better than the alternative, which is to argue while emotionally compromised, which is a slow motion trainwreck
14+
- nobody is going to begrudge you 10 minutes to recompose yourself, they likely want it themselves, and will respect you more for saying it
15+
- I have been in some very heated rooms with attorneys from both sides gaslighting each other
16+
- We seem confused → Are we making progress?
17+
- Manager is rambling → Should we move forward?
18+
- When we’ve unintentionally gone too deep → Can we split the topics please?
19+
- Ok, you have an idea, can you demonstrate it in 5 minutes?
20+
- Is this going to be a bad meeting?
21+
- What’s the goal of this meeting?
22+
-
23+
- Trigger phrases (redirect to powerful phrase)
24+
- This is a waste of time -> We seem confused -> Are we making progress?
25+
-
26+
- Triage
27+
- Is this productive?
28+
- Is this a good use of time?
29+
- What do you expect from me?
30+
- Is this true?
31+
- We’re going in circles, …
32+
- Is this the right thing to do?
33+
- I see that you feel X, I acknowledge that
34+
-
35+
- Concept map
36+
- Wiio’s Laws: [](https://en.wikipedia.org/wiki/Wiio%27s_laws)[https://en.wikipedia.org/wiki/Wiio's_laws](https://en.wikipedia.org/wiki/Wiio's_laws)
37+
- [Amazon writing guidelines](https://www.firstprinciples.ventures/insights/four-tips-to-write-like-amazon)
38+
- “You can’t tell people anything … “
39+
- Curse of development
40+
- Cursed Frame
41+
- [[Aphorisms for Product Management]]
42+
-
43+
- See also
44+
- [[Dealing with difficult people]]

pages/The Feedback Protocol.md

Lines changed: 106 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,106 @@
1+
- This is a framework for product managers conducting user interviews.
2+
- Problem:
3+
- feedback interactions are metastable due to their critical nature
4+
- without a process to guide it, divergence into conflict is the default
5+
-
6+
- The purpose of a feedback framework is to
7+
- make it safe for feedback to occur
8+
logseq.order-list-type:: number
9+
- ensure the feedback is extracted successfully
10+
logseq.order-list-type:: number
11+
- guide the conversation towards convergence and unity of understanding, and away from divergence/conflict
12+
logseq.order-list-type:: number
13+
-
14+
- A healthy feedback frame is marked by two powerful phrases:
15+
- **"Thank you for the feedback"** (vendor -> user)
16+
logseq.order-list-type:: number
17+
- this establishes safety (establishes you the vendor as a safe person) so that a vulnerable conversation can occur. The user is sticking his neck out in criticizing you and your product, they are in a position of vulnerability (and anticipating that you might whack them). We have all experienced failed encounters where feedback was not well received and the relationship was consequently damaged.
18+
- this is for you as much as them; it prompts you into a listening frame of mind and away from the default defensive stance (your biological survival instinct).
19+
- **"That's right"** (user -> vendor) — reflective listening
20+
logseq.order-list-type:: number
21+
- "That's right" is the goal, your job as the interviewer is to get the user to say this
22+
- The phrase serves as a handshake or checksum that the message has been correctly received as opposed to ignored or corrupted in transmission.
23+
- Listen, repeat what you hear, converge gap to zero, until person confirms with “that’s right." It takes more than one loop. At this point the loop has converged to a fixed point. Congratulations, the interview has completed successfully! Say thank you again and that you will use what you learned to inform your roadmap, end the call, you are done!
24+
- (Consider also the opposite, a divergent conversation. If a feedback conversation diverges, the user leaves damaged and will never speak to the vendor again. Analogously, when a marriage diverges, the result is divorce.)
25+
-
26+
- Example of turning around an actively failing conversation:
27+
- [https://clojureverse.org/t/electric-clojure-a-signals-dsl-for-fullstack-web-ui/9788/20?u=dustingetz](https://clojureverse.org/t/electric-clojure-a-signals-dsl-for-fullstack-web-ui/9788/20?u=dustingetz)
28+
- Thank you for the feedback → you’re right → acknowledge → refute central point → answer with numbers
29+
- User reply: “Okay, I see. ... ... Oh wow. By that description of electric, it’s much more thought out than I had first anticipated.”
30+
- [https://news.ycombinator.com/item?id=31219819](https://news.ycombinator.com/item?id=31219819)
31+
- We appreciate your concern -> you’re right → acknowledge -> refute central point with data
32+
- User reply: “This is a fair criticism!”
33+
- (Bomb diffused without detonation! Prospective user feels acknowledged for their contribution! Bravo!)
34+
-
35+
- **"That's right" vs "You're right"** — they are not the same thing
36+
- "That's right" is a listening checksum, it is how you confirm that you have understood correctly which lets the user feel heard
37+
- "You're right" is a magic compliance extraction tool for disarming an antagonist
38+
-
39+
- Anti-patterns
40+
- long responses
41+
- **quote-replies** (instead, **Refute the Central Point**, c.f. Paul Graham [How to Disagree](https://paulgraham.com/disagree.html))
42+
- You don’t need to quote reply and address every little thing. Find the big thing, quote just that, and address it in 3 sentences.
43+
- **fast disagreeable replies** (fast supportive replies are okay)
44+
- people want to feel heard, their experience acknowledged
45+
- over-explaining
46+
- explaining at the wrong level
47+
- counterarguments
48+
- PAUL GRAHAM: "Counterargument is contradiction plus reasoning and/or evidence. When aimed squarely at the original argument, it can be convincing. But unfortunately **it's common for counterarguments to be aimed at something slightly different**. More often than not, two people arguing passionately about something are **actually arguing about two different things**. Sometimes they even agree with one another, but are so caught up in their squabble they don't realize it." https://paulgraham.com/disagree.html
49+
- absolute take on subjective topic
50+
- note: the user is allowed to do this in feedback frame, because you are asking them to express themselves. The vendor is NOT, you must be magnanimous! This is an asymmetric frame!
51+
- avoiding the zoom call
52+
- did you just spend 5 hours thinking about and typing up your rebuttal? How do you know you even rebutted the right thing? With body language, you'd have realized your mistake in the first 10 seconds. If it's not worth a zoom call it's not worth the rebuttal either
53+
- "well actually" https://www.recurse.com/social-rules
54+
- definition: "(conversation) used to introduce a contrary opinion"
55+
- defensive statements and other self-defense responses
56+
- dismissive statements and other expressions of judgement
57+
- the word "No" has no place here, it is a judgement and this is a judgement free zone
58+
-
59+
-
60+
- The problem with longform written communication
61+
- emails, forum posts, inline comments on a document are **terrible mediums for feedback**
62+
- it’s too slow to have error correction feedback loops, inevitably the train goes off the rails and then keeps going a while longer before anyone notices, destroying everything it touches before exploding in a ball of fire
63+
- slack DMs are still too slow ("User is typing a response..."), slack generates negative encounters that could have been positive, and wastes a bunch of time to boot - see [[The Myth Of Async Communication]]
64+
- feedback via body language is immediate and realtime and prevents the train from going off the rails
65+
-
66+
- Vendor is the dominant party, user is the subordinate party
67+
- responsibility for solicitation of feedback, and therefore the frame, is fully on the vendor
68+
- if the user attempts to establish a feedback frame unsolicited, that is unsolicited feedback
69+
- vendor mindset: **magnanimous** ("generous, understanding, tolerant")
70+
- the user is allowed to be confused, the vendor is not
71+
- the user is allowed to be frustrated, the vendor is not
72+
- the user is allowed to express irritation or annoyance, the vendor is not
73+
-
74+
- The feedback protocol does NOT require the vendor to agree with the feedback, only that they **hear** it
75+
- the goal is only to extract and capture the feedback in a way that does not damage the vendor/user relationship (lest the user churn or turn into a detractor)
76+
- the user does not need you to agree with their feedback, **they only want to feel heard**
77+
- analysis is a separate step, it can happen offline.
78+
- It is imperative that analysis DOES NOT occur during the feedback extraction call, because that can lead to a defensive or dismissive tone from the vendor. **If the vendor becomes defensive or dismissive, the feedback frame is immediately poisoned and no further communication can occur.** You have failed!
79+
-
80+
- Analyzing feedback
81+
- During the offline analysis stage, if the vendor doesn't agree with the feedback, that's fine, but be careful: all humans suffer cognitive bias.
82+
- You may not be understanding, or be ready to understand:
83+
- "Communication usually fails, except by accident"
84+
- "What we wish, that we readily believe." — Demosthenes
85+
- Or you may be avoiding or deflecting something that you are not ready to face:
86+
- "Every failure has a moment where you saw it and looked away. That moment is the reason"
87+
- As a rule of thumb, if you hear something three times, there's something there.
88+
-
89+
- How to Disagree? **Irrelevant**, disagreement has no place in feedback frame, it leads to divergence, conflict, discord.
90+
- PAUL GRAHAM: "Even as high as DH5 we still sometimes see deliberate dishonesty, as when someone picks out minor points of an argument and refutes those. Sometimes the spirit in which this is done makes it more of a sophisticated form of ad hominem than actual refutation."
91+
-
92+
- Why feedback matters? Because it generates market learnings -> user alignment -> growth
93+
- The goal of all startups is growth. (The opposite of growth is stagnation.)
94+
- Growth is generated by user alignment.
95+
- User alignment is guided by feedback.
96+
-
97+
- "Crossing the Series A Chasm"
98+
- Most startups get stuck between seed (value hypothesis validation) and Series A (growth hypothesis validation). Often, seed stage founders are building a product with themselves in mind as the Ideal Customer Persona (ICP) and their early adoption comes from innovators who share the founders mindset closely. But the market of innovators is small, and once saturated, growth will stall, because **the needs and expectations of the next concentric circle of adoption are fundamentally different** ([source](https://www.duetpartners.com/crossing-the-chasm-from-seed-to-series-a-2/)).
99+
- This is typically tackled by crafting a [[Positioning statement]] to help bring the product/market/customer into better focus.
100+
- At the seed stage, which we are exiting, we were still in the fog of war with a rapidly evolving hypothesis as we experiment and invalidate various concepts (which segments failed, which succeeded, how can we segment the market on various characteristics in order to isolate the successful use cases where we want to double down).
101+
- Ratcheting the clarify of the position statement and using that to align product roadmap to clearly defined market segments is how you produce the next concentric circle of adoption.
102+
- This is an iterative process that can take years, and is repeated at each order of magnitude of growth (10^3 users, 10^4, 10^5, 10^6, ...)
103+
-
104+
- See also
105+
- [[Powerful Phrases]]
106+
- [[Frames and frame collisions (communication)]]

0 commit comments

Comments
 (0)