Campaigns still send people to knock on doors in an era of precise digital targeting. The practice survives because it performs a function advertising cannot.
Identification comes before persuasion
Before a campaign can mobilise anyone it needs to know who its supporters are, and public records show who voted rather than how they voted.
Field contact fills that gap. A brief conversation at the door produces a recorded assessment of support, which is then used to decide who receives further contact and who does not.
This database is the campaign's most valuable asset by election day, and building it is the main reason field programs begin months before anyone is persuadable.
Turnout is a separate problem from preference
A substantial share of eligible voters already prefer one side and simply do not vote. Persuading them is unnecessary; getting them to the polls is the entire task.
Research on this has consistently found that personal contact outperforms impersonal contact, and that the effect strengthens when the conversation is genuinely two-way rather than scripted.
The practical implication is that a campaign short of money spends it on organisers rather than advertising, because the return per dollar is easier to observe.
Logistics account for most of the work
Voting has friction: registration deadlines, identification requirements, polling place changes, transport and work schedules. Each is a small obstacle capable of stopping a willing voter.
Field programs exist substantially to remove that friction — confirming registration, explaining where and when to vote, and following up with people who requested a mail ballot and have not returned it.
Volunteers are managed rather than merely recruited
A field operation is a temporary organisation that must recruit, train, schedule and retain thousands of unpaid people over a short period.
Attrition is high, so the structure tends to be built around neighbourhood team leaders who hold their own volunteers accountable rather than around a central office issuing shifts.
Campaigns that treat volunteers as interchangeable labour lose them quickly, which is why the organising literature emphasises relationships as much as it emphasises contact counts.
The effect is real but bounded
Field work moves outcomes at the margin rather than transforming them, and the margin only matters in close contests. In a landslide the entire program is invisible.
That bounded effect still justifies the expense, because campaigns cannot know in advance which contests will be close enough for the margin to decide them.