The Futures Wheel: Mapping Consequences of Consequences

A futures wheel is a structured way of working outward from a single change to its direct consequences, and then to the consequences of those consequences. You put the change at the centre, ring the first-order effects around it, then ring the second-order effects around each of those, and keep going. It is not a forecast. It is a disciplined way of finding the implications that a normal planning conversation never reaches, because a normal planning conversation stops at the first ring.

Where it came from

The method was developed in 1971 by the American futurist Jerome C. Glenn, later co-founder of The Millennium Project, as a teaching device for thinking about consequences in an orderly way. It has since become one of the most widely used tools in futures work, documented in Futures Research Methodology and taught in more or less the same form for fifty years.

It belongs to nobody. There is no licence, no proprietary canvas, no certification, and it works on a whiteboard or the back of a printed agenda. We teach it because it is the cheapest possible entry into thinking at more than one order of effect, and because most organisations have never done so systematically.

The rings, and why the first one does not matter much

The structure is simple. The centre holds the change. The first ring holds what follows directly. The second ring holds what follows from each of those, and so on outward.

The value is almost entirely in the second and third rings.

The first ring is usually obvious — already in somebody's slide deck. If your change is "cheap battery storage becomes ubiquitous", the first ring contains "energy bills fall", "grid demand flattens", and everything else the trade press has been saying for three years. Writing it down is necessary, because it is the load-bearing structure for everything further out, but it is not where anything is learned.

The second ring is where it starts, because that is where consequences begin to interact with parts of the world nobody assembled the meeting to discuss. The third is where the genuinely strange things live, and where confidence is thinnest. That tension — the outer rings being simultaneously the most interesting and the least reliable — is the permanent condition of the method, and pretending otherwise is the main way practitioners embarrass themselves.

The historical case that makes this vivid is the motor car. Ring one: people travel further and faster. Ring two: settlement patterns loosen, and the suburb becomes possible. Ring three: the shopping mall, the drive-through, the teenage social life conducted out of parental sight, a national politics organised around the price of petrol. Nobody in 1910 needed a wheel to see ring one. Everything that reshaped the twentieth century sat two and three rings out.

The centre: a signal plus a focus

Most published examples of the method put a vague trend at the centre — "artificial intelligence", "ageing populations" — and the wheel that results is a general essay about the world. It is interesting and it is useless, because it is nobody's problem.

We frame the centre as two things held together: a signal and a focus.

A signal is a specific event that suggests a new possibility has become real. It has a who, a when, and a specific outcome. "AI is getting better at driving" is not a signal. "In October 2020, Waymo opened its fully driverless ride-hailing service to the general public in Phoenix, Arizona" is a signal. The specificity is not pedantry: a vague centre produces vague branches, and a concrete centre forces concrete consequences.

A focus is the domain, organisation, place, or category whose future you are exploring. It is what makes the wheel yours. The same signal with a different focus produces an entirely different wheel, and that is the point — a futures wheel is not a description of the future, it is a description of how one future meets one situation. Where the focus is a place rather than an organisation, the structure holds but the branches run through geography, tenure, and civic life rather than operations; that variant is the spine of our Place Futures work.

Assume it has already happened

Before anything is drawn, we ask the room to assume the signal is fully realised. Not likely. Not imminent. Realised — the barriers overcome, the technology widely adopted, the thing so ordinary that nobody comments on it any more.

This instruction does more work than any other part of the method, and it is the part most often skipped.

Without it, the exercise collapses within ten minutes into a feasibility debate. Somebody says the regulator will never allow it. Somebody else says the unit economics do not work. Both are probably right, and neither observation is what the session is for. Feasibility is a real question, but a different one, and it is the question the organisation is already good at asking. What organisations are bad at is knowing what they would be looking at if the thing did happen.

The assumption is a bracket, not a belief. You are not claiming the future arrives. You are saying: for the next ninety minutes, we will not relitigate whether it arrives, because we want to know what we would be dealing with if it did.

One worked illustration

Take the Waymo signal above. Take as the focus a provider delivering healthcare in people's homes — district nurses, carers, community teams.

Ring one is unsurprising. Staff no longer drive between visits. Travel time is no longer lost time. Parking, which consumes a real and unglamorous share of the working day, stops being a constraint. Fleet costs restructure. None of this would startle anyone who runs such a service.

Ring two is where it becomes a strategy conversation. If travel time is usable time, the vehicle becomes a clinical workspace and the visit stops being the only unit of work. If parking and driving fatigue no longer bound the round, a single carer's catchment widens — which permits teams to be centralised and local depots closed, which quietly erodes the local knowledge that made the service trusted in the first place. If a vehicle can travel without a person in it, some journeys stop needing a person at all: a medication drop, a sample collection. That last one splits the caseload into work that must be done by a human and work that need not be.

That split is the object worth having. Once it exists it becomes a procurement category, commissioners can buy the two halves separately, and the shape of the workforce follows whichever half gets funded. None of that is a claim about what will happen. It is a claim about what would follow — enough to change what the organisation watches for. And notice what did not happen along the way: no argument about whether autonomous vehicles are coming.

How it goes wrong

Four failure modes, in roughly descending order of frequency.

Stopping at ring one. The room fills the first ring, the energy is good, the whiteboard looks productive, and the output is a list the organisation already had. A session that ends with one ring completed has not failed at the futures wheel — it has failed to start one.

Branches that are the same consequence in different words. "Staff drive less", "less time behind the wheel", "reduced driving hours" — three branches, one consequence. This is the most common way a wheel looks full while being thin. The test is deletion: remove a branch and see whether anything is lost.

Losing the causal discipline. Every ring must be a consequence of its parent, not a free association from the centre. If you cannot say the word "because" out loud between a branch and the thing it hangs off, it is not a branch, and the wheel has become a brainstorm wearing a diagram.

Treating the output as findings. A futures wheel produces hypotheses. Nothing on it is evidence. It generates a structured set of if this, then plausibly this claims that still have to be tested — against data, against experts who know the domain, or by building something and seeing how people react. Presenting a wheel to a board as research is how practitioners lose their credibility, usually permanently. The wheel earns its place by being the cheapest way to generate hypotheses worth the cost of testing.

Not a mind map, not a risk register

The futures wheel is routinely mistaken for two things it superficially resembles.

A mind map lets a branch be anything associated with the centre, including a category like "regulation" or "cost". A futures wheel branch has to be something that happens, to somebody, because of the thing it hangs from. That constraint is the whole method.

A risk register works on consequences already identified, scored by likelihood and impact and assigned to an owner. It is a management instrument for known risks inside a bounded programme. The wheel is a discovery instrument for unknown consequences, many of which are opportunities rather than risks, and most of which are too far out to score honestly. Scoring the second ring destroys it: the moment people know a branch will be graded, they write branches they can defend rather than branches they can see. A register can legitimately be an output of a wheel, several steps downstream. It is not a substitute for one.

Where this sits in a larger practice

On its own, a futures wheel is a good ninety minutes. It becomes more than that when the consequences it surfaces are built into scenarios, and the scenarios into specific, concrete future products and services that people can react to — the point at which an organisation stops discussing the future in the abstract and starts making decisions about it.

If you want the whole arc, Speculative Design Basics covers the futures wheel alongside signals, scenarios, and the design work that follows, at your own pace.

Key takeaways

• •A futures wheel maps the direct consequences of a change, then the consequences of those consequences, ring by ring. Jerome Glenn developed it in 1971; it is public domain and needs no special materials.
• •The first ring is usually obvious and already priced into the strategy. The second and third rings are where the value is, and where confidence is weakest — hold both facts at once.
• •Put a specific signal and a focus at the centre. A vague centre produces a general essay about the world rather than anything useful to you.
• •Assume the signal is fully realised before you start. This is what stops the session collapsing into a feasibility debate, which is a real but different question.
• •Every branch must be a consequence of its parent. If you cannot say "because", the wheel has become a brainstorm.
• •The output is hypotheses, not findings. Test them before anyone acts on them.