Trend, passing fad or structural change: how to tell them apart before designing or communicating

A trend usually comes to your attention after someone has already given it a name

A report heralds ‘the next big transformation’. A platform is flooded with the same aesthetic for weeks on end. Several campaigns start using the same visual language, and a presentation sums it up with an upward-pointing arrow. When something has a name, it seems to be understood.

But giving a change a name doesn’t explain how long it will last, how many people have actually adopted it, what need it fulfils, what costs it entails, who stands to benefit from promoting it, or what might hold it back. And those are precisely the questions that matter before turning it into a campaign, a product, a collection or an interface.

This guide proposes a method for answering them. It is not a formula for predicting the future: it is a way of assessing what sort of change you are observing, what evidence is missing, and how much risk you would be taking by acting on it. The analysis begins when you stop asking what the change is called and start investigating what underpins it.

First, describe what you see without using the word ‘trend’

The difference between a trend and a passing fad cannot be determined at first glance, so it is best to avoid letting your conclusion cloud your observation. Instead of stating ‘there is a trend towards X’, describe: certain brands have started to incorporate X; discussion of X has increased on a specific channel; a group of users is using X to solve Y; investments or regulation aimed at X are emerging.

A basic description template:

• What is happening:
• Where is it happening:
• Since when:
• Who is affected:
• How often:
• Which source reports this:
• What we don’t yet know:

The more an observation is qualified by adjectives, the more likely it is to already contain an interpretation. This exercise in precision is the first step in any serious process of trend research.

Don’t investigate the appearance: investigate the mechanism

Manifestations change rapidly; the drivers tend to change more slowly. A particular visual language may be fleeting, but the need that explains it — to communicate in a more accessible, rapid or adaptable way — may be enduring. Aesthetics are the manifestation; the need is the mechanism.

Useful questions to identify it: what need does it address? What cost does it reduce? What new capability makes it possible? What incentive or regulation favours it? Which player is investing in it? What alternative is it displacing? What happens if the novelty wears off?

The evidence dossier: six dimensions a signal must satisfy

If you’re wondering how to identify a trend using more than just intuition, the answer lies in building a case file. Before incorporating a signal into a project, assess it across six dimensions:

An important methodological rule: do not simply add up points as if the result were objective. Assess each dimension as weak, partial, consistent or contradictory evidence. The aim is to acknowledge uncertainty, not to hide it behind a score.

Attention is not adoption

This is the most common confusion when analysing signs of change, and the most costly. Virality measures circulation; it does not prove adoption.

To assess the adoption of trends in a specific market, ask: do people try it or do they keep using it? Who stops using it? Does it become part of their routine? Does it actually replace an alternative? Who pays to maintain it? What happens after the first experience?

Trajectory, infrastructure and behaviour: where to look

Trajectory. Time alone does not validate a hypothesis, but it does allow us to observe its trajectory: emergence, recurrence, speed, expansion, stabilisation or decline. Is it a one-off peak? Does it recur at different times? Does each cycle incorporate new audiences? Has it survived the novelty phase?

Infrastructure. A change gains depth when it ceases to depend on isolated initiatives and begins to become embedded in the functioning of the system: new providers, standards, regulations, platforms, certifications, training, job profiles, and permanent budgets. Infrastructure does not guarantee success, but it indicates that certain actors are turning an expectation into capability.

Behaviour. People and organisations may state that they value something or that they will change their habits. What really counts is something else: repetition, commitment, expenditure, time, persistence, substitution. The key question is: what is someone now doing that they did not do before? It is not enough to consider what they say they will do.

Barriers. Price, access, trust, regulation, skills, habits, productive capacity, cultural resistance. Barriers do not discredit a hypothesis: they allow us to estimate its scope and pace. A well-identified barrier can be more useful for design purposes than an optimistic prediction.

Prioritise your sources and look for what might refute you

A trend supported by a single narrative remains a fragile hypothesis. It is worth distinguishing between three levels:

• Primary evidence: usage data, observable behaviour, regulation, standards, business results, academic research, experiments.
• Secondary evidence: sector reports, consultancy analyses, expert interviews, specialist media.
• Exploratory signals: social media, communities, portfolios, events, shop windows, searches. These help to detect trends; they do not always help to prove them. This is the natural territory of coolhunting, valuable as a radar but limited as evidence.

The rule of triangulation: bring together more than one type of source, more than one stakeholder, more than one context, as well as supporting and contradictory evidence.

Next, subject the signal to cross-examination: what other explanation fits the same data? Who stands to gain from presenting this as a trend? Are we seeing a change or a campaign? Is there a geographical, generational or platform bias? are we confusing investment with demand, availability with utility, early adoption with retention? And the decisive question: what would have to happen for us to recognise that our hypothesis was incorrect? A hypothesis that cannot envisage what would refute it cannot be considered properly evaluated either.

Also watch out for common biases: recency, confirmation, survival, selection, extrapolation from early adopters, and the source’s commercial interests.

The diagnosis is not a definitive label: it is a level of commitment

Following the analysis, draw up a provisional classification:

• A passing fad or superficial phenomenon: intense attention, low retention, little behavioural change, dependence on a single platform, celebrity or novelty. This does not mean it should be ignored: it may be relevant for specific content, campaigns or time windows.
• A trend taking hold: signs across various contexts, growing adoption, repeated behaviour, identifiable mechanisms, and barriers that are beginning to diminish. This may warrant systematic observation, prototypes, pilot schemes or user research.
• Structural change: sustained modification of behaviours, new infrastructure, regulation or standards, long-term investment, multiple sectors affected, and a complex or costly reversal. It may require redesigning processes, acquiring capabilities or reviewing positioning.

The same signal may be structural for one industry, emerging for another and irrelevant for a specific audience. The sector, the geographical area, the time horizon and the quality of the data all influence the assessment.

Not all signals require action: four possible responses

The quality of the analysis is not measured by getting a spectacular prediction right, but by making a decision that is proportionate to the evidence and the cost of being wrong.

Template: the trend dossier

Before designing in response to a signal, write down what would actually change. You can copy this worksheet:

1. Observed signal: what is happening?
2. Hypothesis: what do we think it means?
3. Driving factor: what might be driving it?
4. Audiences affected: for whom does this make a difference?
5. Evidence: what data or sources support this?
6. Barriers: what might hold it back?
7. Timeframe: Over what period might it be significant?
8. Implications: What would change in terms of needs, behaviours, channels, materials, technology or business?
9. Current decision: ignore, monitor, experiment or integrate.
10. Review indicators: what will need to be monitored to update the assessment?

The same signal requires different tests depending on what you are designing

Knowing how to analyse trends also means accepting that the standard of evidence depends on the discipline. Some examples:

• Fashion and fashion communication: is the consumer changing, or is it just the aesthetic discourse? Focus on purchases, repeat business and actual behaviour, not just the rhetoric.
• Advertising and branding: does the signal alter audiences’ perceptions and expectations? Market research and campaign response are key.
• Product and interiors: can it become a viable and widely used solution? Consider materials, costs, manufacturing, regulations and the product lifecycle.
• UX/UI and digital design: is there recurring use, or is it merely curiosity about a new form of interaction? Retention, friction, accessibility and standards are key factors.
• Animation, video games and digital art: does the language transform production and narrative, or is it a temporary stylistic trend? Key factors include the production pipeline, tools, audience and costs.
• Technology, data and AI: does the capability work outside a controlled demonstration? We weigh up performance, cost, scalability, security and regulation.

This is the approach that cuts across the fields of Design, Innovation and Technology: observing, researching, interpreting, prototyping, validating and deciding. It is not about guessing what is coming, but about critically evaluating what is supposedly coming.

A basic vocabulary for discussing the future without exaggerating it

• Signal: an indication of a possible change.
• Weak signal: an early, ambiguous and as yet peripheral manifestation.
• Trend: a sustained and observable pattern of change.
• Fad: a phenomenon attracting a great deal of attention but lacking depth or staying power.
• Driver: a force that promotes or explains change.
• Barrier: a factor that limits adoption.
• Structural change: a transformation of the operating conditions of a system.
• Megatrend: a broad, long-term change affecting multiple sectors.
• Foresight: a structured exploration of possible futures to improve present-day decision-making.
• Horizon scanning: systematic monitoring of emerging signals, risks and opportunities.
• Coolhunting: the detection and interpretation of cultural and consumer signals.
• Scenario: a coherent description of a possible future situation, not a prediction.

If a degree programme mentions trends and innovation, ask how it researches them

If you’re looking into design, communication or technology courses, this method can also serve as a filter. At an open day or an admissions session, ask: what distinction is made between a signal, a trend and a passing fad? How are changes in behaviour researched? Do they work with primary sources and real users? Do they study future scenarios? Do they prototype solutions before committing to a project? How do they avoid bias when interpreting signals?

Don’t just ask which trends you’ll be studying. Ask what method you’ll use to decide which ones matter.

Frequently asked questions

What is the difference between a trend and a fad?

A fad gains attention and superficial adoption quickly, but relies heavily on novelty and is easily abandoned. A trend follows a more sustained trajectory, spreads to new contexts and changes behaviours or decisions. The difference depends not only on how long it lasts, but on the mechanisms that sustain it.

What is a structural change?

It is a transformation that alters the operating conditions of a system: incentives, regulation, infrastructure, capabilities, costs or habits. It can manifest itself through various different trends and is usually more difficult to reverse than an aesthetic preference or a market novelty.

Does going viral prove that something is a trend?

No. Going viral indicates circulation and attention within certain channels. To identify a trend, it is important to look at adoption, repetition, behaviour, investment, infrastructure and longevity. A conversation can grow very quickly without actually changing people’s real decisions.

How can you tell if a trend will last?

It cannot be known with certainty. Its strength can be estimated by analysing its trajectory, adoption, behaviour, infrastructure, barriers and the independence of the sources. It is also advisable to define what evidence would necessitate a review of the hypothesis and which indicators will be monitored over time.

What sources are useful for analysing trends?

It depends on the sector. It is advisable to combine behavioural data, official documents, regulations, academic research, business results, user research, sector reports and cultural indicators. A single consultancy, platform or publication should not form the basis of the entire conclusion.

What is a weak signal?

It is an early, minority or ambiguous indication of a change that has not yet taken hold. It may foreshadow a significant transformation or fade away without consequence. Its value lies in raising a hypothesis to be investigated, not in proving on its own that the change will occur.

When should a brand act on a trend?

When the signal is relevant to its audiences and objectives, there is sufficient evidence, and the cost of experimentation is proportionate. It is not always necessary to commit the entire strategy: some signals warrant observation or reversible pilot schemes before being fully integrated.

What is the relationship between trends and design?

Trends provide context, but they are no substitute for research or judgement. Design must interpret which needs, behaviours or conditions are changing and translate them into useful, viable and coherent decisions. Merely following the outward appearance of a trend without understanding its underlying mechanisms often results in superficial solutions.

Interpret thoroughly before designing

The next time someone presents something as ‘the trend that will change the sector’, don’t start by asking how to apply it. Describe the signal, look for the mechanism, examine adoption, identify the barriers and decide how much risk it warrants. Designing with a forward-looking approach isn’t about being the first to follow, but about interpreting things better.