MAYA: The Right Amount of Weird
Ship the most novelty a market will swallow — one notch past familiar, never past comprehensible.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 93%
Consumers migrate behaviour, they do not reinvent it. Rory Sutherland uses Raymond Loewy's MAYA principle — Most Advanced Yet Acceptable — as a calibration dial: push novelty right up to the edge of what feels acceptable, then stop. Products that are too familiar are invisible; products that are too strange trigger rejection, disbelief, or paranoia. The design target is 'the right amount of weird'.
Origin
MAYA (Most/Maximally Advanced Yet Acceptable) was coined by the French-American industrial designer Raymond Loewy. Sutherland applies it to product, pricing and UX decisions, and illustrates it with the BBC buyer who bought the Danish series The Killing on the theory that Scandinavians are 'just the right amount of weird'.
Core principles
- 01Evolution beats reinvention: even the iPhone was preceded by the iPod, so it was not a complete WTF moment.
- 02Novelty is a dial, not a switch — the goal is maximum advancement subject to an acceptability constraint.
- 03Over-optimisation can cross the acceptability line: a razor made too quiet reads as broken, not better.
- 04Weirdness is only valuable when the audience can still decode it — Denmark is interesting, Somalia is unintelligible, to a Western viewer.
- 05Acceptability is a psychological constraint, not a technical one, so it will not show up in an engineering spec.
How to run it
- 1
Name the behaviour you are asking people to change
Write down the existing habit the product replaces and the exact delta in behaviour you are demanding of the user. The size of that delta, not the size of the technical leap, is what determines adoption risk.
Pro tip Ask: can the user migrate into this from something they already do, or must they reinvent their routine?
- 2
Find the acceptability ceiling
Establish what the market currently accepts as normal in this category, and how far past it a reasonable buyer will tolerate before the product feels alien, suspicious or embarrassing to own.
Pro tip Look for the analogue reference point (iPod → iPhone) that gives the new thing a familiar handhold.
Watch out Reviewers and early press set the acceptability frame — Meta Portal TV was a genuinely great product buried under nine paragraphs of privacy paranoia.
- 3
Push to the ceiling, then stop
Advance the product as far as possible while staying inside the acceptable band. Deliberately leave the last increment of technical improvement on the table if crossing it makes the product unbelievable or uncomfortable.
Watch out Do not confuse 'better on the spec sheet' with 'better in the hand'. The crackle and buzz of a razor is the user's proof that it is working.
- 4
Preserve sensory reassurance cues
Identify the noises, frictions and physical signals users read as evidence the product is doing its job, and keep them even when the technology no longer needs them.
Pro tip If you remove a cue, replace it with a deliberate substitute signal.
- 5
If timing blocks acceptability, wait and re-launch rather than kill
A rejected product may be mistimed rather than wrong. Sutherland's rule: do not reject things simply because they failed in the past — in a complex system, the same action can produce a different result later.
Pro tip Keep a list of intrinsically good products that failed on timing; they are re-launch candidates, not graveyard entries.
In the wild
A BBC Four buyer paid roughly £24,000 for a Danish crime series that had aired in Denmark eighteen months earlier and gone nowhere else. His stated rationale was that Scandinavians are 'just the right amount of weird' — foreign enough to be interesting to a British or American audience, familiar enough that viewers are not lost.
→ The Killing became hugely popular, migrated from BBC Four to BBC Two, and Netflix later bought the rights — Sutherland estimates tens of millions of pounds returned to the Danish broadcaster.
An engineer used a machine-learning approach to make a personal-care device (an electric razor or toothbrush) almost silent. Sutherland warned him that the buzz and crackle are what convince the user the device is working.
→ Sutherland's counsel was to stop short of maximum quietness — a device optimised past the acceptability line reads to the user as not working at all.
Common mistakes
Treating maximum technical optimisation as maximum value
You can over-optimise. Removing the very friction or noise that signals function to the user destroys perceived value even as measured performance improves.
Demanding reinvention instead of migration
Consumers like to migrate their behaviour rather than reinvent it. A product with no familiar handhold forces a psychological leap most of the market will not make, regardless of how good it is.
Writing off a failed category forever
'Doing the same thing again and expecting a different result' is not insanity in a complex system — it is how complex systems behave. Many good ideas failed purely on timing.
Is it for you?
Best for
Founders and product leaders launching a genuinely novel product or a category redesign, deciding how far past the current norm to push.
Not ideal for
Commodity or infrastructure products where the buyer explicitly wants zero novelty and pure spec-sheet performance.
From the transcript
“there is a pace of change which consumers will accept and generally they're more comfortable with Evolution than they are with complete reinvention”
“as I said don't go crazy weird but there's this concept of Maya which is maximally advanced yet acceptable”
“the great thing about Scandinavians is they're just the right amount of weird”
“razor a bit quieter but if you make it too quiet we genuinely won't believe it's working”
From the episode
What most people miss about marketing
Rory Sutherland (Vice Chairman of Ogilvy UK, author)