The AI Backcasting Strategist skill for Claude

You download the zip at the top of this page, put it in your Claude environment, and from that moment on Claude plans backwards. What a skill actually is and how that works technically is explained in what Claude skills are.

You recognise the problem this skill solves by the feeling that a goal is too big to start on. You know where you want to go, somewhere in five or ten years, but every plan you make sets off from today: current customers, current diary, current constraints. The result is a list of resolutions that is just a little more ambitious than last year, and that never arrives at the end goal. The file names that mechanism literally: incremental forecasting keeps people within their current constraints, backcasting makes visible the path that would otherwise stay hidden.

Who is this for? For anyone with an ambitious goal and no path: business owners with a future vision for their company, teams that need to turn a product roadmap or a sustainability plan into actions, and people with a personal goal that collapses every January. Those are the situations the file itself names as the moment to use the skill. It is part of the library we make available for free through the AI and automation service, with no account and no sales email afterwards.

One thing you should know beforehand: the skill does not formulate your future vision for you. That is one of its hard rules. It asks the questions that sharpen the vision, it makes sure every milestone is measurable, but the content comes from you. Anyone who hopes the AI will simply invent the goal gets a follow up question instead.

The difference with forecasting is bigger than it sounds. Forecasting extrapolates: this is where we stand, this is the trend, so that is where we end up. That works fine for the question of how much stock you need next month. It does not work for a goal that requires a different state than today, because every extrapolation takes your current constraints along as a fixed given. Backcasting deliberately parks those constraints. First the future is described richly and concretely, as if you are already there. Only then comes the question of how you get there, and that is answered backwards: what needs to be true just before the target year, what three years earlier, what five years earlier, and what does that mean for this month.

In practice you feel the difference fastest in the language of the plan. A forecast talks in percentages and intentions, a backcast in states you can check. The skill guards that difference with an explicit rule: never advise in growth percentages alone, milestones describe states. Below you see what that does to three typical plan lines, with the milestones from the example in the file as the replacement.

The red lines are hope, the green ones are checkable. That is not a matter of style: a state can be tested on a date and a percentage intention cannot, and that turns the plan into something you can steer by instead of something you have to believe in. Anyone who wants to apply the same reversed direction of thinking per project instead of per year finds the Working Backwards Coach skill in the same library, built on the method Amazon uses to work back from the press release.

Then follows the framework in five phases. Phase one defines the future vision, and the file sets the bar explicitly: a good vision is specific, time bound, desirable, not limited by current resources and imaginable. It asks questions about it out loud: what year is it in the vision, what does an ordinary day look like, which figures belong to it, what do you no longer do then, who is on the team and who are the customers, and what is different about the world, your industry or your region.

Phase two describes the present honestly, with no invented starting point: where do you stand now on those same variables, which resources, skills, network and capital are there, which habits, customers, contracts or identity are fixed, and what is the real constraint instead of the self imposed one.

Phase three makes the gap explicit on five axes: financial, operational, people, market and personal. Per axis not only the difference comes on the table, but also the question of which patterns need to be broken instead of optimised.

Phase four is the heart: the backcast in milestones, from the target year via T-1, T-3 and T-5 back to now, exactly the sequence from FIG.02. Every milestone must describe a measurable state instead of an activity, be logically reasoned back to the previous one, and not conflict with physical or fundamental laws.

Phase five closes with the first steps and blockers: three concrete actions for this month, three patterns that need to be broken, three assumptions to test, three early signals that show the vision is becoming more achievable, and what you can already stop doing now because it does not contribute to the goal.

The file closes with five hard rules, which you see in action in FIG.04 below, and a source list of four titles: Robinson from 1982, Robèrt from 1989, Holmberg and Robèrt from 2000 and Dreborg from 1996. If you want to learn to build a file like this yourself, the full explanation is in writing a SKILL.md.

After that, using it is simple: name your goal and your horizon, or just say your goal feels too big. It starts with the vision questions and holds onto you until every milestone is a measurable state. The step by step explanation with screenshots for every environment is in installing Claude skills, and anyone who wants to understand working with AI more broadly first starts in the knowledge base.

There are also situations where you are better off leaving it alone. For short term work backcasting is too heavy: a quarterly plan or a sprint goal does not need a ten year future vision, and the strength of the method actually grows with the distance to the goal. It is not a crystal ball either: it does not predict whether your vision will come true, it makes the path towards it explicit and testable. Less exciting than a prediction, and considerably more usable. That is why every plan ends with assumptions to test and signals to follow, because a path that is never held up against reality is still just a wish.

And anyone who first wants to know where they even want to go starts a step earlier: choosing a vision is different work from working a vision backwards, and for personal goals with obstacles the WOOP Method Coach skill sits closer.

The most important limit is your own honesty in phase two. The file requires a present with no invented starting point, but it cannot check whether your figures are correct. Anyone who presents their current situation as prettier than it is gets a gap that looks too small and milestones that arrive too early. And it goes without saying that a conversation with an AI does not replace business or financial advice for big decisions: the skill sharpens your thinking, the choices and the responsibility stay yours.

Many future visions contain a line somewhere about customers who come in on their own. That is usually the milestone where the path gets stuck, because findability does not happen by itself. Phase two of this method asks for an honest baseline measurement, and for your online position there is one within seconds: the free SEO scan. What comes out of it is immediately the starting value for the milestones you record afterwards.