The Assumption Nobody Remembers Making

Some of the strongest constraints in an organisation are decisions whose original reasons have disappeared.

Every mature organisation contains rules that nobody remembers introducing. A drawing standard requires a particular detail, a procedure specifies an interval, a purchase specification excludes a material, an approval needs a certain signature. Most began for good reason. The problem is that organisations preserve the rule more reliably than the circumstances that made it sensible.

Circumstances change unevenly. Technology improves, suppliers develop new capabilities, regulations are revised and specialist materials become routine. The procedure remains untouched because nothing about it appears to be failing, and its continued existence becomes evidence of its necessity. A new employee sees that something has been done the same way for twenty years and reasonably assumes that somebody, somewhere, has already tested the alternatives. Usually nobody has.

The longer this continues, the less the assumption looks like a choice. It begins to appear as a property of the system: the customer will not accept it, the regulator requires it, the equipment cannot tolerate it. Any of these statements may be true. The relevant question is whether the person repeating it knows why it is true now, rather than why it was true when the rule was written.

Industrial systems are particularly exposed because physical assets live for decades. A fabrication limitation shapes a design standard, which shapes a maintenance method, which shapes inspection access, which eventually shapes cost and reliability. Years later the fabrication limitation may have disappeared while every downstream arrangement remains, and improvement efforts deliver incremental gains inside a structure that no longer needs its present form. The same happens outside engineering: an approval level set when managers had less information, a reporting requirement introduced after an incident long since resolved, a commercial term that reflects a balance of market power that has since shifted. In each case age is mistaken for proof.

Challenging such assumptions is not a rejection of experience. Experience is valuable precisely because it stores the results of earlier mistakes. It loses value, however, when the lesson is separated from the condition under which it was learned, because a rule without its reasoning cannot distinguish a permanent principle from a temporary workaround. One way to recover the distinction is to ask what would need to be true for the rule to become unnecessary. The question moves attention from defending the practice to identifying what supports it. If the answer is a physical law, a live regulatory requirement or a genuine economic constraint, the boundary is real. If it is a supplier limitation from 2008 or a customer preference nobody has checked since, the organisation has found something worth revisiting.

The hardest assumptions to reach are those attached to identity. When a company defines itself through a product, process or capability, an alternative stops being an option and starts being a threat. Evidence against the assumption is then not merely inconvenient but personal, because accepting it would require the organisation to describe itself differently. Established firms often recognise a technical shift early and respond late for exactly this reason: the analysis is available, but the self-description is not yet ready to change. Assumptions of this kind need a different challenge, one that separates what the organisation does from what it is for.

Useful organisational memory contains more than instructions. It preserves enough reasoning for a future person to judge whether an instruction still belongs to the world in front of them. The aim is not to make every rule temporary, but to stop temporary conditions from acquiring permanent authority simply because nobody remembers that they were temporary.