The PESTEL Analyst skill for Claude

You download the zip at the top of this page, put it in your Claude environment, and from that moment Claude produces environmental analyses following this logic. What a file like this actually is and how it works, you can read in what Claude skills are NL.

A PESTEL analysis maps out the six external forces that fall outside your control but still affect your strategy: political, economic, social, technological, ecological and legal. That last word is Legal, so legislation and law. You also come across the same model as a PESTLE analysis, and that is not a different model: it is the same six factors with the L and the E in a different order in the acronym. Some courses write PESTLE, some books write PESTEL, and there are variants called STEEP or DESTEP. The skill recognises all of those spellings as a trigger, and in every case it delivers the same six dimensions in the same fixed order, so your analyses stay comparable with each other.

The problem the skill solves is the tunnel vision almost every business owner suffers from. You know your own business down to the details, you know your customers, you know your competitors by name. What you do not see is the legislation coming into force in eighteen months, the subsidy scheme closing next quarter, the demographic shift in the region where you are considering opening a branch. It is exactly that kind of force that often decides whether a plan works. Francis Aguilar already wrote in 1967 that managers structurally spend too much time on internal figures and too little on the environment around them, and that observation has never gone out of date.

This skill is meant for anyone who has to build the case for a big decision: business owners considering a new market or region, marketers preparing a launch, consultants who need to paint a picture of a sector, directors writing a board pack or a multi year plan, and investors doing due diligence. Together with the rest of the skill library of 100 free skills, we make it available from the AI and automation service, without an account and without a sales email afterwards.

One thing you need to know beforehand: PESTEL looks exclusively outward. Your own organisation, your team, your processes and your resources are deliberately left out. That is not a shortcoming but the boundary of the model, and the skill actively guards that line. If you want to put the inside and the outside next to each other, the SWOT analysis skill is the logical partner: the opportunities and threats from a PESTEL form exactly the O and the T of a SWOT.

It turns into a trend list. The classic mistake is filling six columns with terms like an ageing population, digitalisation and sustainability, and calling that an analysis. A list like that is not wrong, it is just empty: it does not say what the factor does to your plan. That is why the skill forces four extra fields for every factor. Impact, probability, time horizon and rationale. Without those four it is not a factor but a word.

The analysis reads the same for every sector. This is stated as a literal prohibition in the file: no PESTEL that reads the same for every sector. If you can swap out the company name without the analysis changing, you have filled in a template instead of thinking. That is why the skill asks for the exact scope first, and why the rationale for each factor has to refer to the subject, not to the world in general.

Predictions become facts. The temptation is strong to write down an expectation as if it were certain, especially when the analysis has to support a decision you have really already made. The file explicitly bans that: no predictions dressed up as facts. Where there is no figure or source, the word assumption goes in instead, and that word also comes back in the final section of the output.

The analysis quietly turns inward. As soon as someone writes that the team is too small or that the software is outdated, it is no longer a PESTEL. Those are internal points and belong in a SWOT or a value chain analysis. This is also a hard prohibition in the file, and it is the most common way the model gets contaminated.

No conclusion comes out of it. Six filled dimensions are not yet advice. That is why every analysis from this skill ends with a synthesis in four parts: the top three opportunities, the top three threats, three to five strategic recommendations and an honest list of what you still do not know. The ranking is done on impact times probability, so not on what sounds most exciting.

The difference is clearest in this fictional side-by-side comparison, whose premises and ratings remain unverified. On the left the line as it appears in most environmental analyses, on the right the same line as this skill produces it.

And it also contains situations rather than words: internationalisation plans, a market entry, a product launch in a new region, a five year plan, preparing for a board meeting, an investment decision, due diligence for an acquisition, building the case for a business case, and a strategic reorientation.

After that comes the philosophy, and it is short and concrete. Francis Aguilar published Scanning the Business Environment from Harvard Business School in 1967 and introduced the ETPS framework in it: economic, technical, political, social. Only in the 1980s did Gerry Johnson and Kevan Scholes expand that in Exploring Corporate Strategy into PESTEL, adding the ecological and legal dimensions, because environmental and legal factors were increasingly tipping the scales in strategic choices. The core of the model is summed up in the file in one sentence: strategy is never made in a vacuum.

The heart of the file is the four step method. Step one is the scope: the skill keeps asking until it knows four things, namely the subject, the geographic scope, the time horizon and the purpose of the analysis. If it does not get a clear answer to that, it makes reasonable assumptions and states them explicitly in the output.

Step two is the scan itself: the six dimensions in the fixed order P, E, S, T, E, L, each with its own list of angles to consider. Political has seven, the five other dimensions have eight each, so 47 checkpoints together. They are concrete: for legal it names by name GDPR, the DSA, the AI Act, MiCA and NIS2, plus employment law, competition law, product liability, intellectual property, tax rules, consumer protection and legislation that is still on its way.

Step three is the weighting. A minimum of three and a maximum of seven factors per dimension, and five fields per factor: the factor stated briefly and factually, the impact as positive, negative or neutral, the probability as high, medium or low, the time horizon as now, one to two years or three to five years, and a short rationale for why that factor matters for this subject. Step four is the synthesis: the top three opportunities in order of impact times probability, the top three threats in the same order, three to five strategic recommendations and finally the section What you do not know, with the assumptions, the data gaps and the signals you still need to validate yourself.

After that comes a fully written out output template with all the headings and fields, so the layout does not shift between two analyses, and the fictional padel-hall example in Tilburg, with all premises explicitly marked as unverified assumptions. The file closes with six tone rules and five refusals. The tone rules: write in a business like and factual way, use concrete figures and sources or otherwise label something as an assumption, do not gloss over unwelcome conclusions, avoid jargon or explain it briefly, no emoji, and do not put everything into bullet points. If you want to learn to write an instruction file like this yourself, the method is set out in writing a SKILL.md NL.

It started with four letters. Francis Aguilar published Scanning the Business Environment from Harvard Business School in 1967. His argument was simple and uncomfortable: managers consistently spend too long analysing internal data and too little time looking at the wider environment. His framework was called ETPS, for economic, technical, political and social, and it was meant as a systematic scanning tool, not a fill in the blanks exercise.

The ecological and legal dimensions came later. In the 1980s Gerry Johnson and Kevan Scholes expanded the framework in Exploring Corporate Strategy into PESTEL, because the environment and legislation were increasingly starting to dominate strategic decisions. That is exactly the moment the spelling confusion arose too: PESTEL and PESTLE have both come into circulation and mean the same thing. In both cases the L is Legal, so legislation, and the remaining E is Environmental, so the environment and climate. Anyone who treats the acronym as a memory aid rather than as doctrine has no problem with this.

The weighting is not a given. The source list in the file also names more recent publications, including Yuksel's 2012 work on setting up a multi criteria decision model around PESTEL, and a 2011 case study by Renko and colleagues on applying the five forces model to a small business. The point of those references is that a macro analysis only becomes useful once you weigh the factors against each other. Hence the fixed fields for impact and probability, and hence why the top three are ranked by impact times probability.

PESTEL rarely stands alone. Usually it is the outer layer of a larger analysis package. The opportunities and threats from a PESTEL are the O and the T of a SWOT analysis, and where PESTEL describes the macro environment, Porter’s five forces model NL looks at the structure of your industry itself: new entrants, suppliers, buyers, substitutes and rivalry. The two do not clash, they zoom in at a different level. And anyone who wants to think through the consequences of a macro signal further than the first round ends up at the Second-Order Thinking skill: every measure has an effect, and that effect has an effect of its own.

Even without installing the skill you can use these principles to improve your own environmental analysis: for every factor, note what it does to your plan, how likely it is and when it comes into play, and write down what you still do not know.

After that, using it is simple: name your subject, your region and your horizon, and ask for a PESTEL analysis. The sharper the scope you supply, the less the skill has to assume. If you are new to files like this and want the broader context first, you will find the explanation of working with AI in the knowledge base.

There are also situations where you are better off leaving it alone. The clearest is the internal question: if it is about your team, your processes, your systems or your offering, PESTEL is the wrong instrument, and the skill says so too. A second limit is a very short horizon. Macro factors move over years, not weeks, so for an operational question about next month the model mostly delivers noise. And the third limit is data: the skill works with what you supply and with what it knows, and where that is not enough it writes assumption. For figures that really matter, such as occupancy rates, permit conditions or the size of a local market, validation stays a human job.

One more honest limit: a good environmental analysis gives your decision a better foundation, not automatically a better outcome. If the offering is not right or the financing does not come together, no analysis will change that. What a PESTEL does do is stop you being caught out by something you could have seen coming. Anyone who wants to know, after the analysis, where the plan could still go wrong, follows up with the Pre-Mortem Analyst skill.

A good environmental analysis tells you whether the market you want to enter is moving in your direction. What it does not tell you is whether that market can find you. That is the next step, and that is where our own trade sits. The free SEO scan shows in a few seconds where your site stands. In the guide which strategy model when NL you can see how this model relates to the others. And if you want to talk further about what else AI can do for your business, from separate skills to full automation, we simply do that in a conversation.