tryngumpre

B2B research · claude

Standard

The claude standard is a limit on what you may say

Every guide on this hub inherits one rule. A claim may travel if it names a role, a scene, and a tradeoff, and if a colleague can find those three in the notes. A claim must stay home if it promises ROI or dresses a range of conversations up as representative statistics. That limit is what tryngumpre means by claude.

Read the glossary entry
Researcher reviewing stacks of unmarked notes under an olive lamp
Narrow evidence is not a weaker version of a statistic. It is a different kind of knowledge, and it has to keep its own shape.

Editorial rule for tryngumpre research notes

Claims the desk will carry, and claims it will not

Use the table as an edit pass on interviews, observation writeups, pattern codes, and positioning lines. If a sentence needs the right-hand column to sound impressive, cut the sentence.

Claim shape Carried under claude Refused
What happened In this renewal, the reviewer asked for an exit plan before features were discussed. Buyers always prioritize exit plans over features.
What repeated Across the lost deals in this range, champions stalled when they could not explain rollback. Sixty percent of the market stalls on rollback.
What we will say We describe the tradeoff an operator named, and we name the role. Customers see ROI in the first quarter.
What we did not hear Security did not sit in these interviews. The line does not speak for security. Silence treated as agreement, or filled with an industry benchmark.

How to apply the cut in daily work

  1. 1. Point backward. Every public sentence should point to a note from interviews, observation, or both.
  2. 2. Keep the limit beside the pattern. Pattern synthesis may say what repeated inside the range. It must say who was absent.
  3. 3. Ban outcome theater. ROI, payback months, and “guaranteed lift” do not become acceptable because a buyer hoped for them. Hope is not a measurement.
  4. 4. Ban costume statistics. Words like “most,” “the majority of the market,” and “statistically representative” require a design you did not run. Protocols call the grid a range on purpose.
  5. 5. Let positioning stay narrow. Positioning synthesis fails the standard when the line is only strong if the reader assumes a number.

Friction research remains welcome. A careful account of why a yes stalls, written in purchase friction and checked against a committee map, is more useful than a borrowed benchmark. The standard is strict about costume. It is not hostile to judgment.

A before-and-after edit

A champion says the pilot waited three weeks because security wanted a rollback note, and that the plant manager would not sign until that note existed. The sloppy version becomes “enterprises delay purchases for security.” The claude version stays with the scene: in this pilot, security asked for a written rollback, and the operator withheld a signature until it existed. The second sentence can go into a positioning workshop. The first sentence cannot.

Keep a margin note when you make that cut. Write what you removed and why, especially if the removed line had a number, a superlative, or a promise of savings. Later readers of the glossary should be able to see that “pattern” never meant “market share.” The margin note is also how a new editor learns the standard without a separate training deck. If you cannot point to the scene, the line goes back to the interview file, not out to a homepage.

Field notes

Revisions to the standard, by email

If tryngumpre tightens the claude wording, the note goes to this list. The mail is about evidence hygiene, not about promised commercial outcomes.

Take the standard back into the work

Start again at interviews, or return to the hub and pick the lens your last deck violated.