There is a real science of how to get someone to say yes to a direct request, and it works. It is also the most oversold corner of behavioural science, a place where a handful of robust effects sit beside a pile of charming one-off demonstrations that mostly do not replicate. The applied skill is telling the two apart, and knowing where the ethical line runs.
Compliance is the narrow, practical question of getting a yes to a specific request, and the field's durable contribution is a set of sequential-request techniques that reliably beat a cold ask. Each works by exploiting a different lever.
Foot-in-the-door starts small. Agree to a tiny request (put a small sign in your window) and you become measurably more likely to grant a much larger one later (host a big ugly billboard), because the first yes quietly shifts your self-image toward being the kind of person who helps (Freedman and Fraser, 1966). Door-in-the-face runs the opposite way. Make a large request that will be refused, then follow with the smaller one you actually wanted; the retreat reads as a concession, and the norm of reciprocity pulls the other person to concede in return by saying yes (Cialdini et al., 1975). Low-ball works through commitment: get someone to agree to a deal, then change it for the worse, and a surprising number stick with the choice they have already committed to rather than reopen it (Cialdini et al., 1978).
These techniques are the applied edge of the broader compliance tradition. They draw on the same principles catalogued in Cialdini's six (reciprocity, commitment, social proof, liking, authority, scarcity), treated in detail in L5-13, and at the far end the same machinery shades into obedience to authority, the territory of the Milgram studies covered in L4-15. What distinguishes the material here is that it is a toolkit of moves, not a theory of mind: specific, testable request structures a practitioner can actually deploy.
The honest reckoning is that this literature is a mix of robust phenomena and oversold demonstrations, and the replication crisis has been uneven on it. The broad principles of reciprocity and authority hold up well. Foot-in-the-door survives as a real effect, but meta-analysis shows it is smaller and more conditional than the original demonstration implied (Burger, 1999). And the genre the technique is most famous for, the single-study field demonstration (the towel-reuse sign, the mints with the restaurant bill, the returned hotel keys), has fared badly: many of these celebrated one-off results replicate weakly or only in specific contexts, part of the wider unravelling of headline social-psychology demos (L0).
The techniques also backfire in predictable ways when the target senses manipulation or the setup is wrong, social proof can license the bad behaviour it names, authority fails without perceived legitimacy, scarcity reads as a trick when the constraint looks artificial. Those failure modes are mapped in detail in L5-13. The honest framing for all of it is that Cialdini's principles are plausible families of mechanisms with replicated cores and unreliable edges, not a recipe. Used as hypothesis generators they are valuable; used as guaranteed levers they will disappoint a large share of the time.
Much of the evidence is lab-based or drawn from single field sites, effect sizes swing with context, and the samples skew WEIRD. Reciprocity and obligation norms differ markedly across cultures, and authority rests on locally-conferred legitimacy, so a technique that works in one market or channel travels only as a hypothesis to another, especially across the gap between an in-person ask and a digital one.
Which of these techniques generalise across cultures and channels, and which are artefacts of a particular time and place? Can anyone predict in advance which plausible application will work rather than learning it only by testing? And how do the classic in-person effects change when the request arrives through a screen?
The usable core: compliance techniques are real but oversold, so treat them as hypotheses to test rather than proven levers, lean on the robust cores, and hold a hard ethical line, because the whole toolkit is manipulation-adjacent by design.
The ethics genuinely come first here, more than almost anywhere else in this series, because these techniques are built to move people toward a yes they might not otherwise give. The line that matters is whether the influence would survive the other person seeing exactly how it worked. A genuine small first request, a real concession, an honest scarcity all pass; a low-ball that baits someone with a price you always intended to raise, a manufactured shortage, a fake consensus do not, and beyond the ethics several such moves (bait-and-switch pricing in particular) are simply illegal in many markets. The reliable test for any compliance tactic is the audience one: if it only works because the target does not notice what you are doing, it is a liability dressed as a technique, and the trust it burns costs more than the yes it buys.
Use the techniques as testable hypotheses, not recipes. Foot-in-the-door (a small free trial or a tiny first commitment before the larger ask), door-in-the-face (a premium option that makes the standard one feel reasonable), and reciprocity (a genuine gift or useful thing given first) are all worth trialling, but the evidence says you should expect a substantial share of plausible applications to produce nothing, so A/B test them in your own context rather than trusting the anecdote from a book (Burger, 1999). Lean on the robust cores: reciprocity and legitimate authority replicate more reliably than the colourful one-off tactics. Avoid the low-ball entirely as a customer tactic; it is the clearest bait-and-switch, it corrodes trust the moment it is noticed, and it strays into unlawful territory. The applied reference worth reading with this scepticism in hand is Cialdini's Influence (1984) and Pre-Suasion (2016), and the consultancy built around them, valuable as a catalogue of mechanisms to test, not a set of guarantees.
The ground game already runs on compliance logic: the small pledge or petition signature that makes a later turnout or donation more likely is foot-in-the-door in action, and the commitment techniques behind it are among the better-supported tools (see L5-01 for the turnout evidence). The discipline is the same as everywhere else, treat each tactic as something to test in your actual electorate rather than import wholesale, and hold the honesty line, because manufactured grassroots enthusiasm and manipulative asks read as exactly what they are and can repel the people they were meant to move.
Here compliance is often the literal goal, paying tax, following a regulation, showing up to an appointment, and reciprocity, commitment devices, and simplification are all legitimate tools. Two cautions apply. First, be honest about the evidence: many of the headline one-letter "nudge" wins have replicated inconsistently, so pilot locally and measure rather than assuming an effect will transfer (L0). Second, authority-based compliance depends on perceived legitimacy; an instruction from an institution a community does not trust can fail or backfire regardless of how it is worded, which is a question of standing, not phrasing (L4-15).
Three habits. Treat every compliance technique as a hypothesis to test, expecting a large share to fail, rather than a lever guaranteed to work. Prefer the robust cores (reciprocity, legitimate authority, genuine small commitments) over the colourful one-off tactics that replicate poorly. And apply the honesty test to every tactic: if it works only when the other person cannot see what you are doing, the trust it costs is worth more than the yes it wins.
.png)