What is the KISS principle?
Engineering culture and software practice
The KISS principle holds that a design works best when it is as simple as it can be for its purpose, because complexity is where cost, error and misunderstanding live. It is usually expanded as keep it simple, stupid, and is commonly attributed to Lockheed Skunk Works engineer Kelly Johnson around 1960, though the exact wording and origin are disputed. Softer variants exist, and the phrase is often misused to shut down debate.
What this means
KISS is a plea to resist unnecessary complexity. The more moving parts a design has, the more places there are for something to go wrong, the more it costs to run, and the harder it is for anyone to understand or repair. Keeping a design as simple as its job allows tends to make it cheaper, safer and more robust.
The origin story is a useful picture. The principle is associated with an aircraft designer who reportedly told his engineers that a combat aircraft had to be repairable by an average mechanic in the field, under pressure, with basic tools. Simplicity was not about dumbing things down; it was about making sophisticated machines survivable in the real world.
The word simple carries a lot of weight, and it is easy to confuse with related ideas. Simple is not the same as easy, and it is not the same as simplistic. A design can be hard to arrive at yet simple in structure, and a design can be dangerously oversimplified. KISS is about the first kind of simplicity, the kind you have to work for.
Why it matters
For an organisation, complexity is a running tax. A convoluted process, a tangled system or an over-engineered workflow costs more to operate, breaks in more ways, and takes longer for new people to learn. Simplicity is therefore not an aesthetic preference; it is a maintainability and security strategy, because you cannot secure or reliably run what nobody fully understands.
The principle bites hardest in process and workflow design, and now in AI. The simplest prompt that reliably does the job, with the fewest moving parts, is usually more robust and easier to support than an elaborate chain that impresses in a demo. Vendor demonstrations routinely add features and steps that look valuable but enlarge the surface you must operate; KISS is a lens for asking whether each addition earns its keep.
There is real evidence behind the instinct, though it should be stated with care. Decades of software research link higher complexity, measured in various ways, to higher rates of defects and higher maintenance cost. The relationship is not a simple law, and the measures are imperfect, but the broad direction, more complexity tends to mean more errors and more cost, is well supported.
The catch is that KISS is also one of the most abused slogans in engineering. Because everyone agrees with simplicity in the abstract, the phrase gets used to end arguments rather than win them. A leader who wants simplicity has to argue for it with evidence about cost and risk, not just repeat the acronym.
How it works
Where the term came from
The design motto keep it simple, stupid is most often credited to Clarence "Kelly" Johnson, lead engineer at Lockheed's Skunk Works, the division behind the U-2 and SR-71 aircraft, around 1960. The popular account is that Johnson challenged his engineers to design aircraft that an average mechanic could repair in the field with basic tools. The attribution is not clean, however. Reference works note that the phrase was in use in the US Navy around 1960, that the acronym became widely used only later, and that variants such as keep it simple and straightforward soften the churlish stupid. The link to Johnson rests substantially on Ben Rich's 1994 memoir Skunk Works and on later retellings, and the exact expansion of the acronym is genuinely disputed. The honest position is that KISS emerged in mid-century American engineering and military culture and became attached to Johnson, rather than being a single documented coinage.
Simple, easy and simplistic
The most useful modern refinement of KISS is the distinction between simple and easy, argued by Rich Hickey in his 2011 talk Simple Made Easy. Hickey's point is that easy means familiar or close at hand, while simple means not intertwined, not tangled together. Something can be easy to reach for and yet create enormous complexity, and something genuinely simple can be hard to design. KISS is about pursuing the untangled kind of simplicity, which often takes more effort, not the merely convenient kind. Equally important is the opposite failure: simplistic design that strips away things the problem actually needed. The commonly misquoted line attributed to Einstein, that things should be as simple as possible but no simpler, captures the boundary.
Complexity and defects
KISS sits in a wider family of simplicity ideas: Occam's razor from philosophy, the Unix philosophy of small tools that do one thing well, and formal complexity metrics. The best-known metric is cyclomatic complexity, introduced by Thomas McCabe in a 1976 paper in IEEE Transactions on Software Engineering, which counts the independent paths through a piece of code and is still used to flag code that will be hard to test or maintain. The research since then broadly supports the intuition that greater complexity raises the risk of defects and the cost of change, while cautioning that any single metric is a rough proxy rather than a verdict.
How experienced teams pursue simplicity
Simplicity is treated as work, not luck. The UK government's design principles include the instruction to do the hard work to make it simple, and note that making something simple to use is much harder than making it look simple, especially when the underlying systems are complex. Experienced teams test a proposal by asking what could be removed without losing the purpose, whether each feature or step is used often enough to justify its ongoing cost, and whether a newcomer could understand and repair the thing. They resist adding parts to win a demo, and they treat every new moving piece as something that must be operated forever.
Examples
A professional services firm designs an approval process with seven sign-off stages meant to cover every conceivable case. In practice most requests are routine, the process is slow, and staff route around it. Applying KISS, the firm redesigns to two stages for routine cases and an escalation path for the rare exceptions, doing the harder design work to make the common path simple.
A retailer's team is shown an AI workflow that chains several tools together to draft product descriptions. In testing, a single well-crafted prompt produces comparable results with far fewer failure points and far less to monitor. The team chooses the simpler approach, accepting that the elaborate chain looked more impressive in the demonstration.
A council department inherits a spreadsheet with dozens of interlocking macros that only one now-departed person understood. The complexity is a security and continuity risk because nobody can safely change it. The department rebuilds the calculation as a plain, documented process that an ordinary member of staff can follow and repair, treating simplicity as the maintainable and safer option.
Common misunderstandings
KISS means dumbing things down. It does not. The original aircraft story is about making sophisticated machines repairable under pressure, not about making them less capable. Simplicity is a property of structure, not a reduction in ambition.
Simple means easy. It does not. As Rich Hickey argues, easy means familiar or convenient, while simple means untangled; a genuinely simple design is often hard to reach and takes real effort to produce.
KISS settles arguments by itself. It does not. Because everyone endorses simplicity in the abstract, the phrase is frequently used as a thought-terminating slogan. A real case for simplicity rests on evidence about cost, error rates and maintainability, not on the acronym.
This is just the same as YAGNI. It is not. The existing YAGNI article is a specific rule against building speculative features you do not yet need, and premature optimisation is a specific rule against tuning too early. KISS is the broader simplicity ethos, with roots outside software in aviation and the military, of which those narrower rules are particular applications rather than the whole.
Simpler is always better. It is not. Oversimplification, stripping out something the problem genuinely required, is its own failure mode, captured by the warning that a design should be as simple as possible but no simpler.
Risks and boundaries
The chief risk is that KISS becomes a rhetorical weapon. Used to shut down discussion, it can block necessary complexity as easily as it removes needless complexity, and it can be wielded by whoever wants to avoid work. The folklore also outruns the evidence in places: the origin is genuinely uncertain, the acronym's wording is disputed, and the link between complexity metrics and defects, while real, is a tendency rather than a precise law. Presenting any single complexity number as a verdict overstates what the research supports.
There are clear boundaries. Some problems are irreducibly complex, and forcing false simplicity onto them produces the simplistic failure mode, where important cases are quietly dropped. KISS does not tell you which parts are essential; that judgement requires understanding the problem deeply, which is precisely the hard work the GOV.UK principle refers to. The live debate is less about whether simplicity is good and more about who decides what counts as essential, and how to distinguish a design that is elegantly simple from one that is dangerously incomplete.
What to do next
Ask what can be removed. For any process, system or workflow, make it a habit to ask which steps, features or components could go without losing the purpose, and treat every surviving part as a running cost.
Insist on evidence, not the slogan. When someone invokes KISS, ask them to show the cost, error or maintenance argument behind it, so that simplicity is justified rather than merely asserted.
Design the common path to be simple. Do the harder work of making the routine case straightforward and handling rare exceptions separately, rather than building one elaborate process that serves everyone badly.
Favour the fewest moving parts in AI work. Prefer the simplest prompt or workflow that reliably does the job over an impressive chain with many failure points, and be sceptical of demos that add complexity.
Guard against oversimplification. Check that a simple design has not quietly dropped a case the problem genuinely needed, using the test that it should be as simple as possible but no simpler.
FAQs
What does KISS stand for?
It is usually expanded as keep it simple, stupid, with softer variants such as keep it simple and straightforward. The exact wording is disputed, and the comma before stupid is often dropped in older versions.
Who came up with the KISS principle?
It is commonly attributed to Lockheed Skunk Works engineer Kelly Johnson around 1960, but the phrase was also in US Navy use at the time, and the attribution rests largely on later accounts such as Ben Rich's memoir. The origin is genuinely uncertain.
What is the difference between simple and easy?
In Rich Hickey's framing, easy means familiar or convenient, while simple means untangled and not interwoven. Something easy to reach for can create great complexity, and something truly simple can be hard to design.
Does simpler code really mean fewer bugs?
Research broadly links higher complexity to more defects and higher maintenance cost, and metrics such as McCabe's cyclomatic complexity, from 1976, are used to flag risky code. The relationship is a strong tendency, not an exact law, and single metrics are only rough proxies.
How is KISS different from YAGNI?
YAGNI is a narrow rule against building speculative features before you need them. KISS is the broader simplicity ethos, with origins outside software, of which YAGNI is one specific application.
Can something be too simple?
Yes. Oversimplification, or being simplistic, strips out something the problem genuinely required. The guiding line is that a design should be as simple as possible but no simpler.
Why is KISS so often misused?
Because everyone agrees with simplicity in principle, the phrase gets used to end debate rather than to make a case. Effective use means arguing from cost, risk and maintainability rather than repeating the acronym.