The AI SMART Goals Coach skill for Claude

You probably recognise the problem the skill solves. You write down: more revenue. Or: perform better. Or: live healthier. It feels like a goal, because there is a direction in it. But there is no number in it, no date and no reason, so in three months you cannot say whether it worked. And because you cannot tick it off, it stays on the list until the year is over. Those exact three missing things, the number, the date and the reason, the skill pulls out of your head by asking a question about them.

According to the frontmatter it comes up automatically for sentences like I want more revenue, I want to live healthier or I want to grow, and for the words around them: setting a goal, writing goals, setting a KPI, formulating a target, annual plan, quarterly goal, team goal, personal goal, New Year's resolution, writing an OKR key result, making a resolution concrete, formulating a plan. And, tellingly, also whenever someone simply has a goal with no deadline or no measuring point. This coach belongs to the skill library NL of 100 free skills we make available through our AI and automation service, with no account and no sales email afterwards.

The name is smart-doelen-coach, the version number 1.0.0, it is under our own name and it falls under the MIT licence. The frontmatter states the framework literally: SMART, after Doran, 1981, Management Review. That is the source everything below goes back to.

S for Specific. What exactly do you want to achieve? No vague terms, but a concrete description. More customers is a direction, five new customers in the hospitality sector is a goal. The difference is not in the ambition but in whether someone else can retell what you mean.

M for Measurable. How do you know you have succeeded? A number, a percentage or a demonstrable result. This is the criterion that is missing most often and that delivers the most, because without a measuring point you cannot adjust course halfway. You only find out on the deadline whether it worked, and by then it is too late to do anything about it.

A for Acceptable, elsewhere achievable. Is the goal achievable given your time, capacity and resources? Note the wording: achievable given what you have, not achievable in general. Fifty new customers can be a fine goal for a team of ten and an impossible goal for someone working on it two days a week. That is why the question about your capacity comes before the number is accepted.

R for Relevant. Does this goal contribute to your bigger ambition or strategy? This is the criterion people would rather skip, and precisely the criterion that should scrap most goals. A goal that contributes to nothing is work you are imposing on yourself.

T for Time-bound. When is the deadline? An exact date or period. Not this year, not soon, but 1 October. A goal without a date is a resolution, and a resolution slips.

The methodology goes back to George T. Doran, who wrote an article in Management Review in 1981 titled There is a S.M.A.R.T. way to write management goals and objectives. It states that the idea was developed by Peter Drucker and later refined by Paul J. Meyer. For anyone who wants to follow that trail: Drucker laid the foundation with Management By Objectives in The Practice of Management in 1954, and Meyer described the variant many trainers use today in Attitude Is Everything in 2003.

Step 1, processing the input. You describe your goal in your own words. That is allowed to be vague, and the file itself gives three examples of how vague: more customers, perform better, raise revenue. It then analyses which SMART criteria are already present and which are missing. That is also exactly why you can run it on an existing goal too: you first get a diagnosis, not straight away a rewrite.

Step 2, asking targeted questions. For each missing criterion the skill asks a concrete question. Not all five at once, it explicitly states, but one at a time, in plain English, so you do not get overwhelmed. The file gives four example questions, and they are so everyday that you could ask them yourself: how many new customers do you want exactly, and in what period. How are you going to track whether you are on course. Why is this goal important for you or your business right now. What is the exact date by which you want to achieve this.

Note that the first question fixes two criteria at once, measurable and time-bound, and that the second question does not ask about the outcome but about tracking it.

Step 3, formulating the SMART goal. Based on your answers, the skill writes a fully worked SMART sentence, followed by an explanation per criterion. That is shown here with the hospitality example you already saw above in FIG.02, including the five lines of explanation that go with it. It is deliberately a sentence and not a table: you can carry a sentence over into your plan, not a table.

Step 4, verification and refinement. It then asks whether the formulated goal is correct, so you can adjust before it ends up in your plan. Concrete next steps follow after that, and those return in the output format as the three first steps at the bottom.

That output format is laid out as a fixed block in the file. At the top the heading with your SMART goal, below it the sentence itself, below that the five explanation lines with the letters S, M, A, R and T in front of them, and at the bottom three numbered first steps. Further on, in FIG.04, that block is shown filled in with the example. At the end are the usage moments and the source list of Doran, Drucker and Meyer.

After that, using it is simple: type out your goal the way it sits in your head, however vague, and answer the questions that come back. Expect two to five questions, depending on how much was already in your first sentence. If you are new to this kind of file, start with what Claude skills are NL and with the explanation on installing Claude skills NL. If you would rather write one yourself, the structure is set out in writing a SKILL.md NL. The broader explanation of working with AI is in the knowledge base.

The last two are the most valuable in practice. With a team goal, SMART makes sure five people read the same thing in the same sentence, and that saves the discussion in November about what was actually agreed. And in a plan with fifteen goals in it, it works as a sieve: goals that fail the R test disappear, and what remains is shorter and sharper.

There are also limits. The file names no exclusions, so this is our honest reading after reading the instructions. The main one: the skill makes a goal watertight, it does not decide which goal you should choose. If you do not yet know where you want to go, you get back a neatly formulated wrong direction. For that, look instead at the OGSM Plan Builder skill, which puts goal, strategy and measures together in one coherent whole.

The second limit is scale. This skill works on one goal at a time. If you want to organise a whole period, with a handful of goals that together add up to something, that is the job of the OKR Coach skill. The two complement each other: OKR builds the set, SMART runs each loose Key Result past the five criteria once more. The frontmatter of this skill even names writing an OKR key result explicitly as a reason to call it up.

And then the well-known criticism of SMART itself, which you are entitled to hear honestly: the criterion acceptable can hold ambition back. Anyone who only sets goals that are achievable given current capacity rarely ends up with something that grows that capacity. Against that stands the stretch logic of OKR, which actually encourages setting goals you probably will not fully hit. If you want to know in advance which obstacle you are going to run into, the WOOP Method Coach skill NL is the logical addition: it puts the obstacle and the plan for it next to every goal.

As soon as there is a number in your goal, the question becomes concrete of where that number has to come from. If it is about new customers or new enquiries, the real work sits one level earlier: being found by the people looking for you. The free SEO scan shows in seconds where your site stands. And if you want to talk further about what else AI can do for your business, from loose skills to full automation, we simply do that in a conversation.