Dayparting and geography: real levers, badly used

Time of day and location are among the few targeting levers left that a human still sets. Which means they are usually set once, from assumption, and never revisited.

The dayparting mistake

The common instinct is to concentrate spend during business hours because that is when the phone is answered.

The data frequently disagrees. Research and comparison happen in evenings and at weekends; forms submitted at 9pm are answered at 9am and convert perfectly well. Turning off evenings can remove the highest-intent portion of the week.

The correct approach is dull: pull conversion rate and cost per conversion by hour and by day of week, over a window long enough to be meaningful, and look. Not sessions by hour — conversion rate by hour.

Often the finding is that one specific window is genuinely bad, and the rest is flat. Adjusting for that one window is worth doing; carving up the whole week is not.

The genuine case for dayparting

When response time changes outcomes. In categories where the first responder wins — emergency services, high-competition B2B — a lead generated when nobody is available is worth measurably less.

When budget is tight enough to force a choice. If you cannot cover the whole week, cover the best part of it.

Geography, and the setting that catches everyone

Google's location targeting has two modes. The default — "presence or interest" — shows your ads to people who are interested in your target location, not only people in it.

For a local business this can mean a substantial share of spend goes to people nowhere near you who once searched about your city. Switching to "presence" only is often the single highest-value change in a local account, and it is buried in campaign settings.

Bid adjustments by location

Worth setting where the economics genuinely differ — a city where you have a depot versus one where you do not, or a region where competition is fiercer.

Worth checking with real data first. The geographic report gives conversions and cost by location, and it regularly shows that the area everyone assumed was strongest is merely the largest.