Readiness checks before you sit a full mock: (1) You can map a fresh client letter into features, effects, embodiments, prior-art admissions, and wishes within one reading. (2) You can state, in one sentence each, the closest prior art, the distinguishing feature, and the technical problem it supports. (3) Every claim feature in your draft maps to a disclosure passage. (4) You can explain why your chosen claim category fits the client's commercial need. (5) Each independent claim has a narrower fallback. If any check fails, return to the matching section here rather than drafting another full paper. Administrative details such as enrolment and scheduling are published by the EPO at https://www.epo.org/; a study guide cannot and should not substitute for those pages.
Reading the client letter: extracting the invention from the disclosure
M1 tasks start from a client letter or inventor disclosure. Your first job is to separate what is enabled, what is merely desired, and what the letter asserts but never explains. Draft nothing until that map is written down.
Build the map in four layers: the concrete features of each embodiment, the technical effect attached to each feature, the problem the client believes they solved, and the commercial wishes. A wish such as 'cheap to manufacture' is not a technical effect and cannot anchor a claim or an inventive-step argument unless the disclosure explains how a feature produces a measurable technical result. Effects attached to only one embodiment matter enormously later, because if that embodiment falls to prior art, the effect falls with it.
Exercise: take any patent-style disclosure, for example a sealing assembly with two variants, and annotate it into five buckets: feature, effect, embodiment, prior-art admission, wish. Expected observations when you do this honestly: effects cluster on one embodiment rather than spreading evenly; wishes outnumber enabling statements; and admissions about existing solutions quietly tell you which features you cannot claim as new. Repeating this on three different disclosures should make the pattern obvious before you attempt a full claim set.
Novelty under Article 54: one document, read directly and unambiguously
Novelty is a single-document test. A claim is new only if no one reference, read directly and unambiguously, discloses all its features, including those merely implied. Everything else belongs in inventive step.
Directly and unambiguously is doing heavy work here. A feature need not be spelled out if it is unavoidably present: a document describing a pump with a sealing ring around a rotating shaft implicitly discloses that the ring contacts the shaft. By contrast, common general knowledge cannot patch a missing feature back into a reference for novelty purposes, but it can supply motivation when you later assess obviousness. Keeping these two roles apart in your notes prevents the classic merged comparison in which two references are quietly read as one.
Practice check: take a claim you drafted and one cited document, then list every claim feature against the exact passage that discloses it. Any feature with no passage, explicit or implicit, is a potential novelty hook; every feature with a passage is a feature you must either delete or defend on inventive step. Do this as a table, not prose, and you will see within minutes whether your independent claim has any novelty basis at all.
The problem-solution approach: deriving a technical problem that fits the features
The problem-solution approach runs: identify the closest prior art, list distinguishing features, derive the objective technical problem from the effects those features achieve, then assess obviousness. Each effect must trace to a claimed feature.
Two disciplines keep this method honest in practice. First, the objective technical problem must be derivable from the original disclosure: if the letter never suggests the feature produces the effect you now rely on, the argument rests on hindsight, and an examiner can attack it as such. Second, the problem must not be phrased as a résumé of the solution. Compare 'providing an improved seal' with 'reducing leakage at high pressure using a chamfered groove'. The first invites any reference about better seals; the second anchors a specific feature to a specific effect and gives the combination of references something it must genuinely overcome.
Micro-example: suppose the letter states that test pieces with a chamfered groove leaked measurably less at elevated pressure. Your argument can then define the problem around pressure leakage, because the disclosure supports the link between chamfer and effect. If the letter had said only that the groove 'performs better', a narrower problem built from a claim feature without stated support would be a guess. Match the certainty of your problem statement to the certainty of the disclosure, and say in your answer where the support comes from.
Choosing claim categories: product, process, use, and fallback structure
Different claim categories cover different commercial acts. Choose categories the client's plans require and the disclosure supports, and remember that a use claim can rescue patentability where a product per se is already known.
European practice recognises independent claims in different categories, for example product, process, apparatus, and use, with unity requirements tying them together. The client letter is your guide: a client who manufactures a device needs product claims; a client whose invention lies in how the device is made needs process claims; a client using a known compound in a new technical or therapeutic application may need a purpose-limited use claim, because the compound itself adds nothing new. Drafting every category 'just in case' without support produces claims that add no coverage and multiply attack surface.
A frequent drafting mistake is three independent claims that recite the same core feature in three grammatical costumes. They neither broaden protection nor provide a fallback, and they can raise unity questions. The better structure is one broad independent claim carrying the effect-bearing feature, followed by narrower dependents that add combinations disclosed in secondary embodiments. Use the table below as a first-pass decision aid, then check each row against what the disclosure actually enables before committing.
| What the letter tells you | Category to weigh | Key point |
|---|---|---|
| Client manufactures a physical device carrying the effect | Product claim | Cover the structure that produces the effect; avoid purely result-oriented wording |
| The invention lies in a method of making | Process claim | Recite the steps that matter, not only the machine inputs |
| Known compound applied to a new therapeutic use | Purpose-limited use claim | The product per se lacks novelty; the new use supplies it |
| Client wants breadth before any product is finalised | Broad independent plus narrower fallback | Narrower claims preserve grant chances if the broad one falls |
Worked scenario 1: when prior art destroys the letter's favourite embodiment
When a cited document discloses one of the client's embodiments, stop drafting around it. Re-baseline the analysis on the surviving embodiment, move the technical problem to the effect only that embodiment achieves, and rebuild claims there.
Scenario: the letter describes a lip seal with (i) a PTFE liner praised for material savings and (ii) a 30-degree chamfered groove that test data shows reduces leakage at 200 bar. A cited document discloses a PTFE-lined lip seal with a square groove. The plausible mistake: the practice answer leads with the liner as the inventive contribution because the client wrote about it most enthusiastically, so every claim reciting the liner alone collapses against the reference, and the draft has no novelty basis left where it counts.
The better decision is structural, not stylistic: name the chamfer as the distinguishing feature, define the objective technical problem as reducing leakage at high pressure, and draft the independent claim around the groove geometry, with the liner appearing only in a dependent claim where it combines with the chamfer. Why it matters: the claim and the inventive-step argument now stand or fall together, the disclosure's test data supports exactly the claim you rely on, and a fallback structure survives even if the effect itself is challenged in the argument.
Worked scenario 2: an inventive-step argument that proves too much
An obviousness attack succeeds when your own technical problem is broader than the distinguishing feature supports. Anchor each effect to a claimed feature and confirm the disclosure states or implies that effect before relying on it.
Scenario: a second cited document discloses chamfered grooves on hydraulic cylinder ends, explained as making machining easier. The plausible mistake: the practice answer defines the problem as 'improving sealing performance generally' and argues the chamfer achieves it. That broad formulation invites the combination of references, because easier machining is an obvious reason to use a chamfer anywhere, and the argument has now supplied the motivation link the second reference alone never had.
The better decision narrows the frame: the objective technical problem is reducing leakage at high pressure in a lip seal. The second reference addresses machining convenience in a different assembly and gives no hint about high-pressure leakage, so combining it with the first reference requires a motivation it does not provide. If the letter contains the 200-bar test data, cite it as support for the effect. Why it matters: the argument's strength comes from the fit between feature, effect, and problem, and a candidate who states the problem loosely hands the examiner a bridge that a tighter formulation would have denied.
A preparation sequence and self-check rubric for M1 practice
Build practice around full mock tasks in stages: map the disclosure, write a decision log, draft claims, then write the patentability argument. Score each stage against a rubric before starting the next mock.
An adaptable sequence: first, two sessions doing disclosure mapping only, using any patent-style letters you can construct from published patent documents; second, drafting claim sets from those same letters with the decision log written before any claim language; third, writing inventive-step arguments against documents you supply yourself; fourth, combining all three into complete mocks under a time limit you set to mirror your own working pace; finally, a review cycle in which you re-attempt one earlier mock cold. Notice which stage produces the most rubric failures, because that stage, not drafting speed, is usually where the next study session belongs.
Self-check rubric, scored 0 to 2 per criterion: the decision log names the closest prior art, distinguishing features, and technical problem before drafting begins; every claim feature maps to a disclosure passage; every effect cited in the argument is anchored to both a feature and a disclosure passage; claim categories match the client's stated commercial needs; each independent claim has a narrower fallback. A total of 10 or more out of 12 across two consecutive mocks is a reasonable learning milestone for moving to harder materials. Treat it as a study signal only; no rubric score predicts an examination result.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.