Read every clue before selecting anything
Begin with a complete pass through the grid. Notice unusually specific terms, repeated word forms, possible phrases and clues that could belong to several categories. Do not submit the first pair you recognise: two related clues are evidence, not yet a complete group.
Mark broad possibilities in plain language. For example, ‘things with keys’ is more useful than ‘music’ because it gives you a property to test. At this stage, keep labels provisional and avoid bending a clue to protect an attractive idea.
Build a claim that fits every member
A proposed group needs a sentence that applies to all of its clues in the same way. Use a frame such as ‘Each can follow …’, ‘All are types of …’ or ‘Each contains …’. If one clue needs a pun while the other three use literal meanings, the claim may be too loose.
Write or say the label before selecting the fourth clue. A label such as ‘animals’ may be technically true but too broad to distinguish the intended set. ‘Flightless birds’ or ‘animals appearing in these four familiar phrases’ is easier to test against the remaining grid.
Use a holding area for flexible clues
Some clues are deliberately plausible in more than one group. Put these flexible items in a mental holding area instead of assigning them immediately. First look for a group containing three specific clues and only one plausible completion.
Suppose a clue can name a colour, a surname or a food. Ask which interpretation creates the most exact four-item claim and leaves the remaining clues capable of forming complete groups. The right use of an ambiguous clue often becomes clearer from what it leaves behind.
Test the remainder before committing
Before submitting a group, remove its four clues in your mind and inspect what remains. Can you see at least one workable pair or category among the leftovers? If the proposed group consumes the only obvious member of another strong set, pause and compare the two explanations.
This remainder test is especially useful when five clues appear to fit one broad theme. The intended four often share a narrower relationship, while the fifth is essential elsewhere. Prefer the grouping that produces two clean future claims over one broad group and a collection of leftovers.
Recover from an incorrect group
An incorrect attempt still provides information. Keep any three clues that share a precise label, then reconsider the fourth. Ask whether another unselected clue fits the exact wording more naturally, not merely whether it is also associated with the topic.
If several attempts fail, reset the claim rather than rotating random fourth choices. Split a vague category into possible mechanisms: synonyms, words following the same prefix, members of a named set, titles or phrases. A changed label creates a new test; a random swap usually does not.
Review the explanation after the solve
When the answer is revealed, compare its label with the one you used. Identify the clue that created the most overlap and the word in the intended explanation that made the group exact. This turns a difficult grid into a reusable lesson about how connections are constructed.
A different answer can be interesting without being equally strong. Check whether it groups all clues exactly once, uses one consistent relationship and leaves no weaker remainder. If two full partitions appear equally defensible, use the Fun IQ contact route so the ambiguity can be reviewed against the published puzzle standard.