Nobody lies. The truth just doesn't survive the trip up.
What happens to one frontline truth as it climbs a 500-person company, level by level. Each level passes up most of what it knows, not all of it. The loss compounds. Every number below comes from published research, or is labeled as an assumption you can change.
Theoretical model. The 70% pass-through per level is the midpoint of laboratory findings that 54–77% of available information gets passed upward. What's left at each level is the pass rate multiplied by itself once per hop.
Three things to look at
- 01 · The bottom
Start with one person
About 6,200 thought shifts a day. Minds wander nearly half the time. Then each person decides, mostly without noticing, what to say out loud. What they say becomes the front line's 100%.
- 02 · Each hop
Every level has a reason to soften it
Fear of looking negative. Protecting a relationship. "Nothing will change anyway." None of it is lying. At 70% per level, 70% of 70% of 70% of 70% is 24%.
- 03 · The controls
Move the slider
Small changes in candor at each level make a big difference at the top. Switch to nine levels and the same habits leave the CEO with about 6% of the picture.
The pipes leak inside every silo, and between them
Give the same 500-person company twelve functions. Now truth leaks two ways: up through the levels inside a function, and sideways at every handoff where two functions have to work together. Whatever leaks ends up in the same place.
Illustration, not to scale. Every joint on a pipe is a level; every cross-pipe is a handoff between two functions. Each one drips.
A forecast is this cascade with a dollar sign on it
A commit deal is a frontline truth on its way up. The seller rounds it up a little. The manager rounds it up a little more. By the time it reaches the board deck, the deal that was "probably next quarter" is in this quarter's number, and nobody lied to get it there.
The people closest to the customer hold the most truth and have the least power to move it. Carrying it costs them too: research links the kind of silence where people give up speaking with burnout.
- The forecast holds until week 10, then slips
- Deals die late at a finance or procurement step nobody mentioned
- New reps take far longer to ramp than the plan assumed
- The board deck explains the miss as timing, again
Your people aren't the leak. The joints are.
How the forecast call runs, how handoffs work, what happens to the person who brings bad news. Fix the joints and the truth reaches the top on its own.
The Denial Audit finds where your cascade leaks
No one can measure a company's pass rate directly. What you can measure is the gap: how leadership and the field score the same five places, checked against your own forecasts, pipeline and ramp data, and what people say when their manager isn't in the room.
In three weeks you get where denial lives first, what it cost last year, and a 90-day plan to fix the joints. One flat fee of $25,000.
- Up to 25 people take the Denial Assessment, from the CEO to your newest seller
- Up to 20 confidential interviews
- Every finding backed by one quote and one number
- A 90-minute readout with a 90-day plan
What the research supports, and what it doesn't
Some of these numbers are solid. Some are popular and can't be traced. Both are labeled.
| Claim | Number | Strength | Source |
|---|---|---|---|
| Thoughts per day | About 6,200 thought transitions | Peer-reviewed | Tseng & Poppenk, Nature Communications, 2020 |
| "60,000 thoughts a day, 80% negative, 95% repeated" | No traceable study | Myth | The claimed 2005 NSF source can't be found |
| Mind wandering | 46.9% of waking time | Peer-reviewed | Killingsworth & Gilbert, Science, 2010 |
| Employees who've withheld an important concern from their boss | 85%. Reasons: fear of being labeled, damaging relationships, futility | Small study | Milliken, Morrison & Hewlin, 2003 (40 interviews) |
| Information passed upward to a supervisor | 54–77% (lab studies) | Lab studies | PMI review of Boyles & Wicker (1977), O'Reilly & Roberts (1974) |
| Bias against passing bad news up | Present even when people trust their boss | Peer-reviewed | O'Reilly (1978), in the same PMI review |
| Projects where people can't raise the real issues | About 1 in 8 people share their full concerns and feel heard | Industry survey | VitalSmarts & The Concours Group, "Silence Fails," 2006 |
| Cross-functional teams that are dysfunctional | About 75% fail at least 3 of 5 measures | Field study | Tabrizi, Stanford; 95 teams in 25 companies |
| "Executives see 4% of problems" | The iceberg of ignorance | Unverified | Credited to Yoshida, 1989; no original source found |
| Secrets people carry | About 13 on average; returning to them in your mind is what wears people down | Peer-reviewed | Slepian, Columbia Business School |
| Silence and burnout | Fearful silence linked to burnout, through low psychological safety | Small study | Health Psychology and Behavioral Medicine, 2023 |
| Cost of depression and anxiety at work | About 12 billion working days a year, globally | WHO / ILO | WHO and ILO, 2022 |
What's research and what's model. The per-level pass rate, the mind-wandering rate and the silence findings are measured. Multiplying them up a hierarchy is my model, not a finding: no study has measured compounding loss across five or nine real levels. Three numbers are assumptions: half of a person's thoughts happen during work hours, every hop loses at the same rate, and a decision coming down loses at the same rate as a problem going up.
No reliable number exists for what share of self-talk is negative, or how many thoughts are about work. If someone quotes one, ask where it came from. That's the point.