KeepsCounselAn editorial exercise from KeepsCounsel for private reflection. The examples below are fictional. This is not a therapy tool, a shared couples feature, a diagnosis method or a way to monitor other people.
1. Start with your side of the record
Choose a small set of notes about one kind of interaction: a week of family calls, three conversations with a friend, or several moments with a colleague. Write down what you recorded yourself seeing, hearing, doing and feeling.
Keep the review anchored in your experience. “I felt dismissed after Alex changed the subject” is reviewable. “Alex does not respect me” may be a possibility to examine, but the journal note alone does not prove Alex’s motive or character.
2. Separate observations, interpretations and missing information
Use three labels. Observation: what the note says happened. Interpretation: the meaning you gave it. Missing information: what the note cannot tell you. This protects the useful signal without pretending you can read someone else’s mind.
Relationship reviews can involve strong feelings. Include those feelings honestly, but do not convert them into certainty about another person’s inner life. If a detail is absent, mark it unknown.
3. Look for patterns across people and settings
A pattern may show up with a partner, parent, friend, roommate, manager or teammate. The point is not to decide who is “the problem.” The point is to see which situations repeat, what you did in them, and what questions are worth carrying forward.
Look for counterexamples on purpose. Did the same person respond differently another day? Did a different person create a similar feeling? Did your own expectations, timing or stress level change the interaction?
A worked example: three fictional notes
- Tuesday with a friend: “Maya replied two days later. I felt unimportant and sent a short answer back.”
- Thursday with a sibling: “Jon called while I was working. I answered immediately, felt interrupted, and sounded annoyed.”
- Saturday with a colleague: “Priya asked for a deadline change. I said I could not answer until checking my workload. We agreed to talk Monday.”
A quick conclusion would be, “People do not respect my time.” The notes support something narrower: delayed replies and interruptions affected the writer’s sense of consideration, while the deadline-change note shows a different response to a request. They do not prove what Maya, Jon or Priya intended.
| Relationship context | What I recorded | My interpretation | Counterexample or missing information | Next question |
|---|---|---|---|---|
| Friend reply | Maya replied after two days; I felt unimportant; I answered briefly. | Maybe I read slow replies as rejection. | I do not know why Maya replied late or whether she expected a quick reply from me. | What happens if I ask about response expectations before assuming meaning? |
| Sibling call | Jon called during work; I answered immediately; I felt interrupted and sounded annoyed. | Maybe I treat calls as urgent even when they are not. | Jon may not have known I was working. I chose to answer. | What boundary would make work calls easier to handle? |
| Colleague deadline | Priya requested a change; I asked for time to check my workload; we agreed to revisit Monday. | I made room to check before committing. | This was a work request with an explicit deadline, unlike the other notes. | Where else might a pause before answering help? |
A cautious summary is: “In these notes, response timing and interruptions affected how considered I felt. In the deadline note, I made room to check before committing. I need more information before deciding what anyone else meant.”
4. Copy a worksheet for your own notes
Copy this table into a document or recreate it in a notebook. Blank cells are fine. “Unknown” is better than inventing a motive.
| Relationship context | What I recorded | My interpretation | Counterexample or missing information | Next question |
|---|---|---|---|---|
| One interaction | ||||
| Similar interaction | ||||
| Exception or different setting |
5. Turn the review into one respectful question
End with a question you can test in your own conduct or clarify in conversation. “What boundary do I want around work calls?” is safer than “Why does Jon ignore my time?” “What do Maya and I expect about reply speed?” is clearer than “Does Maya care?”
If you decide to talk with someone, treat the review as your preparation, not as evidence to prosecute them. Use your notes to describe your experience and ask for clarity. Do not share private journal material unless you genuinely want to.
6. Review positive connection too
Relationship notes are not only for conflict or boundaries. Review what went well: a preference someone explicitly shared, a supportive interaction, a reliable ritual, or a meaningful occasion you want to remember.
Keep the same discipline. “Sam said they like quiet birthday dinners” is a stated preference. “Sam hates parties” is your assumption unless they said it. “My aunt checked in after my exam and I felt supported” records a connection without turning it into a permanent claim about what she always does.
A useful positive-review question is: “What did this person actually say or do that mattered to me, and how can I remember it without turning it into a demand or a guess?”
Questions for reviewing relationships
- What did I directly observe, and what did I infer?
- Which part is about my experience, my boundary or my choice?
- What would I need to ask before claiming to know the other person’s motive?
- Does this happen with one person, several people or one setting?
- What counterexample would make my current interpretation less certain?
- What is the smallest respectful question or boundary to carry forward?
- What positive preference, support or occasion did someone explicitly share or show, and what part is still my assumption?
Optional: ask AI to organise, not judge
You can ask an AI tool to sort selected notes into observations, interpretations, missing information and next questions. Share only material you are comfortable sending to that provider. Do not ask it to diagnose a person, decide who is right or infer hidden motives.
Using only these selected journal notes, separate what I recorded from what I inferred. Mark missing information as unknown. Suggest one relationship pattern to examine, one counterexample, one explicitly positive connection to remember, and one respectful next question. Do not diagnose anyone or infer motives.Check every quoted detail against the original notes. A confident paragraph can still overstate what the record supports.
Reviewing relationships is a companion to reviewing yourself.
See how chosen journal memory works in KeepsCounsel.
Read the privacy explanation before sharing personal notes.