Review log template: what a reviewer objected to and what you changed

Five columns: date, who objected, their words, what changed, what did not and why. The artifact I read first when somebody sends me their work.

Five columns and a page of your own scars. It takes half an hour to keep and it is the artifact I read first when somebody sends me their work.

How to use this file

• Anything in [square brackets] is instruction. Delete it as you fill the row in.
• Write the objection down in the reviewer's words, before you answer it. If you write it down afterwards, from memory, you will soften it. Everybody softens it.
• Every row needs a consequence: what changed, or what did not change and why. An objection with no consequence is a note that you were told something.
• Keep at least one row where you held your ground. A log where every objection was accepted says you have no position, which is a worse signal than being wrong once.
• Date every row, and write the date out in full (6 April 2026), the same way as the rest of the pack. One format, held everywhere.
• The worked example below is the review log from the returns case published on analify.com: the same five rows, the same two reviewers and the same dates as section 6 of that case, because a case with two different review logs has not made up its mind. That case is self-directed and built from public sources. The two reviewers are peers who read the document as a favour, not people inside the company, and nobody at that retailer was asked anything at any point.

The log

[Keep the cells short. Two sentences each is plenty. When an entry needs more room, give it a numbered block underneath the table and put "see note 3" in the cell.]

Reviewed version: v1.2, 9 April 2026. Reviewers: two peers, two sessions, neither with access to the retailer. Objections raised: 5. Accepted: 4. Held with reason: 1.

[That summary line takes ten seconds to write and answers the first question a reader has about the log.]

Notes, where a row needed more room

Note 3 (6 April 2026, four in ten). The stronger version of the objection was to cut the number entirely and say only that the branch exists. I kept the count. A reader who wants to check it can repeat my selection in twenty minutes, and a labelled count with a stated method is more useful than a shrug. If the label is doing its job, keeping the number costs nothing; if the label is not doing its job, cutting the number would not have saved me either. The disagreement stands, it is a reasonable one, and this note is where a future reader finds out that somebody competent thought I was wrong.

[Use these notes sparingly. Two or three in a log, not eight. A log that turns into an essay stops being scannable, and scannable is the entire point.]

How to get the review in the first place

You need one competent person and one favour. Two people from different sides of the work, as in the example above, is better: the one who reads models catches different things from the one who reads registers.

Ask for the objections, not the compliment. "Tell me what you would push back on in the first two minutes" produces something usable. "What do you think?" produces "looks good", which costs you a favour and teaches you nothing.

Then keep quiet while they talk. The urge to explain what you meant is enormous, and every explanation you give in the room is a sentence that should have been in the document. If you have to explain it, that is the finding.

If you genuinely have nobody to ask, say so and do the honest substitute: run the scorecard against your own document two days after you wrote it, and log what it catches with your own name in the "who objected" column, marked as a self-review. It is weaker than a second opinion and everybody knows it. It is still worth more than a document nobody has read twice, and it beats inventing a reviewer, which is a thing people do and which reads exactly as badly as it sounds.

Why this is the strongest document in your portfolio

A polished final draft is cheap. Anybody can produce one, from a template, in an afternoon, and most of what I am sent looks like that: every heading present, every sentence smooth, nothing in it that had to be earned. It tells a reader that you can format a document. It does not tell them what you do when somebody who knows more than you says the thing is wrong.

The review log answers that, and it is close to impossible to fake convincingly. Faking it means inventing objections against yourself, and invented objections are always soft. They are the ones you already knew how to answer. A real one lands somewhere you were not looking, and the entry that shows it is the one where the change is awkward, or where you had to go back and build an artifact that did not exist yet.

Then there is the last column, which is the reason the table has five and not four. A log where every objection was accepted shows compliance. A log with one line saying "I did not change this, and here is the reason" shows judgement, which is the thing being hired. You can be wrong in that column and still come out ahead of the person with no column at all.

Put it second in your portfolio, right after the one-page summary. Judgement at the top, diligence underneath.

What a weak log looks like

• Objections with no consequence. Somebody said something, and the document is exactly as it was.
• Every objection accepted. Either the reviewer was gentle or you did not push back once, and neither reads well.
• Objections that flatter the author. "He said it was very thorough but suggested a clearer heading" is not a review, it is a reference.
• Paraphrase everywhere. Real quotes are short, blunt and slightly rude. Smooth ones are reconstructions, and a reader can tell.
• No dates and no versions, so nobody can see whether the document moved before or after the objection.

Scorecard checks that apply to this artifact: 19 Every entry names who raised the objection, what changed as a result, and the date 20 At least one entry records an objection you did not accept, with the reason you held your ground 21 Every objection is either closed or carries an open action with an owner and a due date

Template 5 of the Proof Pack, from analify.com. Use it in your own applications and at work, including commercially. Do not resell it as your own product.

In the worked case

Section 6 of the worked case shows this artifact filled to the standard the template asks for: returns and refunds at a mid-size online retailer, built from public sources only, with every number tagged. Read it there, then come back to the blank.

The scorecard checks for this artifact

A short record of what a reviewer objected to, what you changed, and what you did not.

All twenty-five checks, with the scoring scale: the scorecard.

Notes that work on this artifact

Next template

All six templates · The worked case · The scorecard