These are the seven signs every article lists. Each one has a far more common innocent explanation, and acting on them wrecks more relationships than it saves. Behaviour-change checklists have poor predictive value for infidelity. The behaviours listed are common, the base rate of cheating in any given month is low, and the arithmetic means the overwhelming majority of people showing these signs are not cheating on anybody
Sudden Phone Secrecy

The number one entry on every list. It is also what happens when somebody is job hunting, managing a health worry, planning a surprise, dealing with a family problem, or has simply started using a privacy screen. Phone behaviour changed for everybody over the last decade.
New Grooming Habits

Gym membership, new clothes, more attention to grooming. Presented as classic evidence and equally consistent with a birthday approaching, a health scare, a work promotion, or somebody deciding they had let things slide.
Working Late More Often

Genuinely ambiguous and almost always explained by work. Industries have peaks, projects overrun and job insecurity makes people stay later. Treating it as evidence tends to punish the partner who is under the most pressure.
Being Less Available

Reduced contact and slower replies are among the most cited signs and among the least specific. Depression, burnout, grief, stress and simple busyness all produce exactly this, and several of them need support rather than suspicion.
Unexplained Spending

Occasionally meaningful and frequently not. Gifts, debt somebody is embarrassed about, a hobby they expect to be criticised for, and financial trouble being concealed all look identical on a statement.
Accusing You Instead

Sometimes a deflection and sometimes a person who is genuinely insecure. The pattern is real enough to be worth noticing, and it is nowhere near reliable enough to conclude anything from on its own.
Being Unusually Attentive

The entry that exposes the whole genre. A partner becoming more affectionate appears on these lists as guilt behaviour, which means both increased and decreased attention are treated as evidence. No test that accepts every result is a test.
What Actually Predicts It

The research points somewhere else entirely. Previous infidelity is the strongest single predictor, with roughly 45 per cent of people who cheated once doing so again against about 18 per cent of those who never had. After that come low commitment, relationship dissatisfaction and opportunity.
Gender Predicts Little

Worth stating because the genre is heavily gendered. Long-running survey data shows a modest gap, driven almost entirely by older age groups. Among adults under thirty, women and men report infidelity at close to the same rate, with women fractionally higher.
Surveillance Has A Cost

Checking a partner’s phone and monitoring their movements is consistently associated with worse relationship outcomes whether or not anything is found. The behaviour itself corrodes the trust it claims to be protecting.
The Better Question

Whether somebody is cheating is largely unanswerable from behavioural clues. Whether a relationship is in trouble is entirely answerable, by asking about commitment, satisfaction and how conflict gets handled. That conversation is available and the checklist is not.
If Suspicion Persists

Persistent suspicion is itself a problem worth addressing, regardless of whether it turns out to be founded. Relationship counselling services exist for exactly this, and the conversation generally goes better than people expect it to.