---
title: Write Down Your Hiring Predictions. Check Them at Day 90
description: Hiring calls are predictions nobody checks. In a 1972 experiment, 84% recalled a higher forecast than they gave. Write three, then check at days 30, 60 and 90.
image: https://nobusato.jp/hubfs/covers-2026-10/hiring-predictions-90-day-check-cover-en-2026-10.png
---

[Nobu Sato](https://nobusato.jp/en?hsLang=en)

[Profile](https://nobusato.jp/en/profile?hsLang=en) [Method](https://nobusato.jp/en/blog/method?hsLang=en) [Articles](https://nobusato.jp/en/blog?hsLang=en) [Glossary](https://nobusato.jp/en/blog/glossary?hsLang=en) [Contact](https://nobusato.jp/en/contact?hsLang=en)

[日本語](https://nobusato.jp/ja/blog/hiring-predictions-90-day-check)[English](https://nobusato.jp/en/blog/hiring-predictions-90-day-check)[中文](https://nobusato.jp/zh/blog/hiring-predictions-90-day-check)

Organizations & Hiring

In this article

1. [Does interviewing more make you a better judge?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s1)
2. [What happens to a prediction you didn't write down?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s2)
3. [What a checkable prediction looks like](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s3)
4. [Day 30, day 60, day 90](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s4)
5. [Whose answers are being checked?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s5)

1. [Home](https://nobusato.jp/en?hsLang=en)
2. [Articles](https://nobusato.jp/en/blog?hsLang=en)
3. [Organizations & Hiring](https://nobusato.jp/en/blog/tag/organizations-hiring)

# Write Down Your Hiring Predictions. Check Them at Day 90

Memory edits the call you made

[Nobu Sato](https://nobusato.jp/en/profile?hsLang=en)7 min read

Contents5

1. [Does interviewing more make you a better judge?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s1)
2. [What happens to a prediction you didn't write down?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s2)
3. [What a checkable prediction looks like](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s3)
4. [Day 30, day 60, day 90](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s4)
5. [Whose answers are being checked?](https://nobusato.jp/en/blog/hiring-predictions-90-day-check#s5)

Managers new to Japan tend to treat the three-month probation period as the check on a hire. It checks the employee. Nothing in it checks the people who made the call, and they are the ones who will make the next ten. A hiring decision is a prediction, and most predictions made in hiring meetings are never written down, so nobody can later say whether they were right. Two old findings from psychology explain why that matters more than it sounds.

## Does interviewing more make you a better judge?

Only if you find out whether you were right. In hiring you mostly don't.

Daniel Kahneman spent his career showing how intuition fails. Gary Klein spent his showing how the intuition of fire commanders and nurses works. In 2009 they wrote a paper together in American Psychologist, subtitled "A failure to disagree." Their shared position: to judge whether an intuition can be trusted, look at how predictable the environment is and at how much opportunity the person has had to learn its regularities. They add that subjective experience is not a reliable indicator of judgment accuracy. Feeling sure tells you little.

A fire commander learns within minutes whether a call was right. An interviewer who says "she'll be fine" in a hiring meeting waits months. By then the outcome is mixed with how the manager handled the first weeks, how the product did and what the market did. And what "fine" meant was never recorded.

Experience under those conditions builds fluency and confidence. It doesn't have to build accuracy.

## What happens to a prediction you didn't write down?

You remember having made a better one. The longer you wait, the more it improves.

The classic experiment is from 1972. Before President Nixon's trips to China and the Soviet Union, Baruch Fischhoff and Ruth Beyth asked students at two Israeli universities to put a probability on each of fifteen possible outcomes of each trip. After the trips they handed out the same questionnaire and asked the students to fill in the numbers they had given before. If they couldn't remember, they were to give the number they would have given at the time.

For outcomes the students believed had happened, about three quarters wrote down a higher probability than their original one. In the groups asked after about two weeks it was 67%. In the groups asked after three to six months it was 84%.

Asked three to six months later, 84% remembered having rated the outcome as more likely

Before Nixon's 1972 trips to China and the USSR, students at two Israeli universities gave probabilities for fifteen possible outcomes of each trip, then were asked afterward to reproduce them. Values are the share whose reproduced probability was higher than the original, for outcomes they believed had occurred. Fischhoff and Beyth (1974 technical report; published 1975 in Organizational Behavior and Human Performance)

- Asked after about two weeks

  67%
- Asked after three to six monthsday 90 falls here

  84%

By the time the result is visible, memory has already edited the original call.

Figure 1. In Fischhoff and Beyth's experiment, for outcomes people believed had occurred, 67% reproduced a higher probability than their original when asked after about two weeks, and 84% when asked after three to six months.

Now put that in a hiring meeting. The new hire is doing well, and the interviewer remembers being impressed from the start. The new hire is struggling, and the same interviewer remembers having had doubts. Both memories are sincere. A review of past hires run from memory tends to conclude that everyone was roughly right.

So the answer key has to exist before the result does.

## What a checkable prediction looks like

I'd write each one as: by what date, what will this person be doing on their own, and who will be able to see it. I'd stop at three. Past that, nobody rereads the page. Three and ninety are my own choices. I have no data showing they are the right numbers.

The uncheckable kind is easy to recognize. "Will ramp up fast." "Shows ownership." "Can lead a team." On day 90 each of these can be scored as correct whatever happened.

The checkable kind is a visible action with a date. Take a head of sales. Day 30: can describe the current customer mix in numbers and say where deals stall. Day 60: has put a quarterly plan on the table under their own name, with the people and budget they need. Day 90: at least one new channel has reached a trial order. After each line I'd add half a sentence on what I would have to see to call the prediction wrong.

The more senior the role, the more I'd put the first line on understanding and plan, and the less on results. My view is that in the first month the right leader should produce an understanding of where things stand and a plan for what comes next, and both should be read at the level of numbers.

I'd add one column that the method page doesn't have: what the company promised. Budget, headcount, authority, how fast head office answers. These are supplied by the environment and are no part of the person. The [four layers](https://nobusato.jp/en/blog/method-four-layers?hsLang=en) keep the pairing of person and environment as a layer of its own, Context, for the same reason. When a prediction misses, I'd read this column first. If the conditions were never delivered, what was misjudged was the environment, and that is logged separately from misjudging the person.

## Day 30, day 60, day 90

The page goes to the hiring manager before the start date. It is read on day 30, day 60 and day 90, and on day 90 it is held against the original verdict.

The glossary calls this [first-90-day verification](https://nobusato.jp/en/blog/glossary?hsLang=en#first-90-days). Which of the [five levels](https://nobusato.jp/en/blog/method-fit-check?hsLang=en) was the person placed at? Which layer was judged weakest? Do the signals at 90 days agree? Ideally someone with no stake in the hire reads it too. Managers grade their own hires gently.

If a new fact comes in along the way, I'd redo the verdict. I'd keep the old one and add the new one as a line below it. Deleting the old verdict lets memory do its editing again.

## Whose answers are being checked?

The hiring side's. The page tests whether the call was accurate. It is not there to decide what happens to the new hire.

That is why I'd show it to the person on day one. It says what is expected by when, and what the company has committed to in return. Probation in Japan usually runs about three months, and [it doesn't mean you can simply let someone go](https://nobusato.jp/en/blog/probation-dismissal-in-japan?hsLang=en). Keeping the predictions hidden and producing them on day 90 like a report card is unfair and also useless. Expectations and rules belong on the table before the start date.

After a year I'd expect the pages to start paying. Ten hires in, it should be possible to count whose predictions held, who runs optimistic, and whether they misread Equipment or Context. The next time two interviewers disagree in a hiring meeting, there would be a record to weigh them by. That part is a guess. I have no data for it.

The candidate is keeping books too. Careful people turn what they were told about scope, budget and headcount into [their own 90-day list](https://nobusato.jp/en/blog/first-90-days-verification-list?hsLang=en). Two private lists are worse than one shared page.

Probation checks the employee. This page checks the people who hired them.

Sources (as of November 2026)

- [Kahneman & Klein, "Conditions for intuitive expertise: A failure to disagree," American Psychologist 64(6), 2009](https://doi.org/10.1037/a0016755)
- [Fischhoff & Beyth, "'I Knew It Would Happen': Remembered Probabilities of Once-Future Things," Oregon Research Institute technical report, 1974](https://apps.dtic.mil/sti/tr/pdf/AD0780971.pdf)
- [Same study, Organizational Behavior and Human Performance 13(1), 1975](https://doi.org/10.1016/0030-5073(75)90002-1)

---

Author

[Nobu Sato](https://nobusato.jp/en/profile?hsLang=en)

Launching in Japan, starting with people and systems. I write about building organizations, sales teams and technology organizations in Japan, in English, Japanese and Chinese.

[Profile](https://nobusato.jp/en/profile?hsLang=en)[Contact](https://nobusato.jp/en/contact?hsLang=en)

## More in this column

1. 01[How to Score Modesty in Japanese Candidates](https://nobusato.jp/en/blog/scoring-modesty-in-japanese-candidates?hsLang=en)7 min
   
   Western samples inflate self-assessments by 0.87 standard deviations; East Asian samples by zero. Scored literally, ...
2. 02[Real Failure or Safe Failure? Reading the Interview Answer](https://nobusato.jp/en/blog/real-failure-or-safe-failure?hsLang=en)8 min
   
   Over 90% of student job candidates fake to some degree in interviews. To tell a real failure from a safe one, ask what ...
3. 03[Why a Good Recruiter Sometimes Tells You Not to Move](https://nobusato.jp/en/blog/why-good-recruiters-say-stay?hsLang=en)7 min
   
   Recruiters in Japan are paid on the hire, yet good ones sometimes say stay. Refunds, public six-month leaver counts and ...

[Column guide: Organizations & Hiring](https://nobusato.jp/en/blog/guide-organizations-and-hiring?hsLang=en) · [Glossary](https://nobusato.jp/en/blog/glossary?hsLang=en) · [The method behind the articles](https://nobusato.jp/en/blog/method?hsLang=en)

[← All articles](https://nobusato.jp/en/blog?hsLang=en)

## Writing about launching in Japan, in three languages.

[Profile](https://nobusato.jp/en/profile?hsLang=en) [Get in touch](https://nobusato.jp/en/contact?hsLang=en)

Nobu Sato

Writing about launching in Japan, in three languages.

[info@nobusato.jp](mailto:info@nobusato.jp) [Substack](https://nobusato.substack.com)[Newsletter](https://www.linkedin.com/newsletters/the-japan-entry-memo-7519750783555518464/)[Medium](https://medium.com/@nobusato5888)[LinkedIn](https://www.linkedin.com/in/realnobusato/)[X](https://x.com/nobusato5888) [RSS](https://nobusato.jp/en/blog/rss.xml?hsLang=en)

[Profile](https://nobusato.jp/en/profile?hsLang=en)[Method](https://nobusato.jp/en/blog/method?hsLang=en)[Articles](https://nobusato.jp/en/blog?hsLang=en)[Glossary](https://nobusato.jp/en/blog/glossary?hsLang=en)[Contact](https://nobusato.jp/en/contact?hsLang=en)

Views here are my own and don’t represent any organization I work or have worked for.

© 2026 Nobu Sato. All rights reserved.·[Privacy and disclaimer](https://nobusato.jp/en/privacy?hsLang=en)

```json
{
  "@context" : "https://schema.org",
  "@type" : "BreadcrumbList",
  "itemListElement" : [ {
    "@type" : "ListItem",
    "item" : "https://nobusato.jp/en",
    "name" : "Home",
    "position" : 1
  }, {
    "@type" : "ListItem",
    "item" : "https://nobusato.jp/en/blog",
    "name" : "Articles",
    "position" : 2
  }, {
    "@type" : "ListItem",
    "item" : "https://nobusato.jp/en/blog/tag/organizations-hiring",
    "name" : "Organizations & Hiring",
    "position" : 3
  }, {
    "@type" : "ListItem",
    "item" : "https://nobusato.jp/en/blog/hiring-predictions-90-day-check",
    "name" : "Write Down Your Hiring Predictions. Check Them at Day 90",
    "position" : 4
  } ]
}
```

```json
{
  "@context" : "https://schema.org",
  "@id" : "https://nobusato.jp/en/blog/hiring-predictions-90-day-check",
  "@type" : "WebPage",
  "about" : {
    "@type" : "Thing",
    "name" : "Organizations & Hiring"
  },
  "author" : {
    "@id" : "https://nobusato.jp/#person"
  },
  "description" : "Hiring calls are predictions nobody checks. In a 1972 experiment, 84% recalled a higher forecast than they gave. Write three, then check at days 30, 60 and 90.",
  "inLanguage" : "en",
  "isPartOf" : {
    "@id" : "https://nobusato.jp/en/blog",
    "@type" : "Blog",
    "name" : "Articles"
  },
  "keywords" : "Organizations & Hiring",
  "name" : "Write Down Your Hiring Predictions. Check Them at Day 90",
  "url" : "https://nobusato.jp/en/blog/hiring-predictions-90-day-check"
}
```

```json
{
  "@context" : "https://schema.org",
  "@id" : "https://nobusato.jp/#person",
  "@type" : "Person",
  "alternateName" : [ "佐藤伸明", "Nobu Sato", "左滕" ],
  "description" : "Japan market launches: organization, sales and technology",
  "knowsAbout" : [ "Japan market entry", "Organization design", "Talent acquisition", "Sales organization building", "Technology organization building", "Japanese labor law" ],
  "name" : "Nobu Sato",
  "sameAs" : [ "https://www.linkedin.com/in/realnobusato/", "https://x.com/nobusato5888" ],
  "url" : "https://nobusato.jp/en/profile"
}
```

```json
{
  "@context" : "https://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Nobu Sato",
    "url" : "https://nobusato.jp/en/blog/author/nobu-sato"
  },
  "dateModified" : "2026-11-12T05:57:10.380Z",
  "datePublished" : "2026-11-12T05:57:10.000Z",
  "headline" : "Write Down Your Hiring Predictions. Check Them at Day 90",
  "image" : [ "https://nobusato.jp/hubfs/covers-2026-10/hiring-predictions-90-day-check-cover-en-2026-10.png" ],
  "mainEntityOfPage" : {
    "@id" : "https://nobusato.jp/en/blog/hiring-predictions-90-day-check",
    "@type" : "WebPage"
  },
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "url" : "https://nobusato.jp/hubfs/seo/logo-nobusato-2026-10.png"
    },
    "name" : "nobusato.jp"
  }
}
```