Dayparting in Google Ads: The Complete 2026 Guide (with Heatmap)

Dayparting is one of those Google Ads topics where the top search results were written when manual CPC ruled the world. The advice was fine in 2014. In 2026, most of it quietly does nothing — because on most Smart Bidding strategies, Google ignores your ad schedule bid adjustments entirely.

This guide is the version we wish existed: what dayparting is, what actually still works under Smart Bidding, how to read hour-by-day data without fooling yourself, and how to act on it — including the honest caveats.

What is dayparting in Google Ads?

Dayparting (also called ad scheduling) means controlling when your Google Ads run — by day of week and hour of day — instead of serving 24/7. In practice it covers three levers: turning ads off during certain hours, raising or lowering bids during certain hours, and shifting budget toward the hours that convert.

Google’s native tool for this is the ad schedule: per-campaign rules like “Monday–Friday, 08:00–18:00” with an optional bid adjustment on each time block (for example, +20% during business hours, −30% overnight).

That’s the definition. Now the part most guides skip.

The 2026 reality check: Smart Bidding ignores your hourly bid adjustments

Here is the single most important fact about dayparting today, and the one the stale guides miss:

If your campaign runs on Target CPA, Target ROAS, Maximize Conversions, or Maximize Conversion Value, Google’s Smart Bidding does not use your ad schedule bid adjustments. You can set +30% for Tuesday mornings and it changes nothing. The algorithm already bids per-auction using time of day as one of its signals, and it disregards your manual hourly modifiers.

What is always honored — on every bidding strategy — is the schedule itself: the on/off windows. If your ad schedule says “no ads on Sunday,” no ads run on Sunday, Smart Bidding or not.

So the honest strategy-by-strategy breakdown looks like this (as of mid-2026 — Google changes this kind of thing, so verify against current documentation before betting the account on it):

Two footnotes worth knowing:

• Enhanced CPC is gone. Google removed eCPC for Search and Display campaigns in 2025 and migrated those campaigns to Manual CPC. If an older guide tells you to “use eCPC to keep some control,” it’s describing a strategy that no longer exists.
• This is why the old advice fails silently. Nothing errors. Nothing warns you in the interface when you set a schedule bid adjustment on a Target CPA campaign. The modifier just sits there, decorative, while you believe you’re dayparting.

Since the majority of accounts now run conversion-based Smart Bidding on most spend, the majority of “set hourly bid adjustments” advice on the internet is advice to do nothing. Everything below is built on that premise.

What still works in 2026

Dayparting isn’t dead — the bid-adjustment version of it is mostly dead. Four things still genuinely work.

1. Ad schedule on/off windows

Turning ads off during hours you never want is honored by every strategy, including Smart Bidding.

When it’s right: you literally cannot serve the click. A call-only campaign when nobody answers the phone. A same-day service business that can’t dispatch at 3 a.m. Lead forms your sales team only works during business hours and speed-to-lead is the whole game.

Pros: absolute, reliable, works on every strategy, instantly reversible.

Cons — and these are real:

• An off window is a hard cap. You’re not lowering bids in weak hours; you’re buying zero impressions. If demand exists at 9 p.m., a competitor gets it.
• Smart Bidding learns from data. Cutting hours cuts the data it learns from, and restricting when it can serve gives the algorithm fewer auctions to optimize across. Google explicitly discourages tightly constraining Smart Bidding for exactly this reason.
• Once an hour is dark, you get no more data from it — ever. You can’t detect that Sunday evenings started converting in November if you switched Sundays off in June. If you do cut hours, schedule a re-test window every quarter or two.

2. Bid adjustments — on the strategies that still honor them

If you run Manual CPC (or Maximize Clicks), classic dayparting still works exactly as the 2014 playbooks describe: find your strong and weak hours, set percentage bid adjustments per time block, review monthly.

The catch is scope: this only applies to the shrinking share of campaigns on those strategies. It’s a legitimate tool for manual-bid shops, small accounts that never fed Smart Bidding enough conversions, and advertisers who deliberately keep manual control. It is not a tool for the typical tCPA/tROAS account — there, it’s a placebo.

3. Budget and campaign-split strategies

If you can’t tell Smart Bidding how to bid by hour, you can still control how much money each daypart gets — by splitting the campaign.

The pattern: duplicate a campaign, give each copy a mutually exclusive ad schedule (say, business hours vs. evenings/weekends), and assign each its own budget and its own CPA/ROAS target. Now “weekend traffic gets a stricter target and a third of the budget” is enforceable, because targets and budgets live at the campaign (or portfolio) level.

Honest cons: you’ve split your conversion data across two campaigns, which weakens each one’s Smart Bidding learning; you’ve doubled the management surface; and small accounts usually don’t have the volume to feed two learners. This is a big-spender pattern. If the daypart in question is less than, say, a fifth of your spend, the split usually costs more in lost learning than it gains in control. Most accounts are better served by option 4.

4. Auditing Smart Bidding with hourly data

This is the modern job of dayparting data, and almost nobody frames it this way: hour-by-day data is how you check whether the algorithm’s time-of-day behavior is actually sane.

Smart Bidding bids by time of day using its own signals. Usually it’s decent at it. Sometimes it isn’t — it will happily keep spending overnight on traffic that never converts if the auction prices look attractive, and it optimizes toward its conversion definition, which may not match where your margin is. A 7×24 view of cost, conversions, and CPA is the audit: if 2–5 a.m. has burned meaningful money over 60–90 days with nothing to show for it, that’s not “the algorithm knows best,” that’s a finding — and an on/off window (or a schedule-split with a stricter target) is a defensible response.

Even when you change nothing, knowing where the money goes by hour is worth having. Wasted spend hides in the dark corners of the week the same way it hides in search terms you never review.

How to read hourly data properly

Most dayparting mistakes aren’t strategy mistakes — they’re reading mistakes. Four rules.

Day-of-week × hour, not hour-of-day averages

An “average performance by hour” chart lies to you. Monday 9 a.m. and Saturday 9 a.m. are different auctions, different users, different intent — averaging them into one “9 a.m.” number blends a strong cell and a weak cell into a mediocre one. Always read the full 7×24 grid: seven days × twenty-four hours, 168 cells. The pattern you can act on (“weekday mornings strong, weekend mornings dead”) only exists in the grid.

Respect minimum data thresholds

168 cells means your data is spread thin. Thirty days of a mid-sized campaign might be 4–5 observations per cell. A cell with 6 clicks and 0 conversions is not a finding; it’s noise.

Rules of thumb we use before letting a cell drive a decision:

• At least ~30 clicks and ~$25 of spend in the cell before “no conversions here” means anything.
• At least ~3 conversions in the cell before you compare its CPA or ROAS to the campaign average.
• For zero-conversion calls, sanity-check against the campaign’s own conversion rate: if the campaign converts at 5%, a 10-click cell with zero conversions is entirely expected (about a 60% chance by pure luck). A 60-click cell with zero conversions is a real signal.

If a cell doesn’t clear the bar, widen the date range or aggregate up (all weekday mornings together) instead of acting on the cell.

Watch the seasonality traps

A 30-day window that contains Black Friday, a sale, a holiday, or your competitor’s stockout is not a picture of “your hourly pattern” — it’s a picture of that event. Before acting, compare the window against the previous period: if “Thursday evenings are amazing” only appears in the promo month, it’s the promo, not the Thursday. Longer windows (60–90 days) smooth one-offs but can also blur a real recent shift, so check both.

Account for conversion lag — it makes recent evenings look worse than they are

Google Ads reports conversions at the time of the click, not the time of the conversion. That’s actually good for dayparting — a Tuesday 9 p.m. click that converts Wednesday morning is correctly credited to Tuesday 9 p.m. But it creates a trap: the most recent days in your window haven’t finished accruing conversions yet. If your typical click-to-conversion lag is 2–3 days, the last few days of any date range will under-report conversions — and if your window ends today, the freshest cells look uniformly worse than they are. Judge conversion metrics on a window that ends several days back, or mentally discount the trailing edge.

And one more: everything is in your Google Ads account’s timezone. More on that in the mistakes section, because it burns people constantly.

Step-by-step: building an ad schedule in Google Ads

The UI moves around, so treat the click-path as “as of mid-2026” and the concepts as stable:

1. Open your campaign, and find Ad schedule — currently under the campaign’s Audiences, keywords, and content section in the left navigation.
2. Click to add a schedule and pick the campaign.
3. Build your time blocks: day (or day-range) plus start and end hour. Google allows a limited number of schedule blocks per day (six, in the current product), so group hours into contiguous ranges rather than 24 one-hour slices.
4. Critical: any hours you leave out of the schedule are hours your ads will not run. If you only want bid adjustments, not blackouts, cover the full week with blocks first.
5. After saving, the Ad schedule table shows your blocks with a Bid adj. column — set percentage adjustments per block there (remembering everything above: on tCPA/tROAS/Max Conversions/Max Conversion Value, these adjustments are ignored).
6. Times run in the account timezone, and the timezone effectively can’t be changed later — double-check it before scheduling anything.

Then wait. A schedule change needs weeks of data before you can judge it, not days — see the thresholds above.

The heatmap shortcut

Everything in the “how to read it” section is work you can do by exporting hourly segments into a spreadsheet and pivoting. We got tired of doing that, so we built it into Keyword Ninja: the Dayparting Heatmap.

What it actually is (no more, no less):

• A 7×24 grid — day of week × hour of day — of your synced Google Ads performance, for any date range, filterable by account and campaign, with hour-of-day and day-of-week rollup views.
• Eleven switchable metrics: cost, clicks, impressions, conversions, conversion value, CTR, CPC, conversion rate, CPA, ROAS, and value per conversion. Ratio metrics are computed from summed components per cell — never averaged averages.
• Period comparison (this month’s grid vs. the previous period), a conversion-lag warning on fresh data, your account’s timezone displayed on the page, and CSV export if you’d rather do your own analysis.
• If you’ve connected GA4, an optional engagement layer shows site engagement by hour alongside the ads data.

The grid is included on every plan, including the free one.

On Pro and up, it goes further than read-only:

• Recommendations score every campaign × day × hour cell against that campaign’s own baseline — flagging slots that spend with zero conversions, slots with a CPA far worse than the campaign average, and slots that convert well enough to deserve more. Every recommendation is gated behind minimum-sample rules (the ~30 clicks / ~$25 / ~3 conversions thresholds above, plus a statistical check on zero-conversion slots), so a six-click cell can never trigger a bid suggestion.
• One-click apply can push a full-week ad schedule with bid modifiers to Google Ads — with guardrails we’ll state plainly: it previews as a dry run with an estimated-savings figure before touching anything; it only writes to campaigns that have no existing ad schedule (so undo means “remove exactly what we added,” available from your activity log); it always writes a complete week so your serving stays 24/7; and if the campaign runs Smart Bidding, it refuses to apply and tells you why, giving you step-by-step manual guidance instead of silently setting modifiers Google would ignore. We built the reality-check from this article into the button.

If you just want to see your own grid before deciding anything, the free Google Ads audit is the fastest way in — connect an account, get the analysis, no card.

Common dayparting mistakes

Reacting to low-volume cells. The classic. Someone sees “$40 spent Wednesday 2 p.m., zero conversions” on 12 clicks and blocks the hour. Twelve clicks at a 4% conversion rate produces zero conversions most of the time by pure chance. Thresholds first, decisions second.

Cutting hours that assist conversions. Click-time attribution helps hourly analysis, but journeys are messier than one click: people research on their phone at 10 p.m. and come back to convert through a branded search — or another device — the next day. Evening and weekend hours often feed the weekday cells that look great. Before blacking out an hour, check whether overall conversions dip after the change, not just whether that cell’s direct numbers looked bad. When in doubt, bid down (where it works) or set a stricter split-campaign target instead of going dark.

Timezone confusion. Ad schedules and hourly reports run in the account’s timezone. A New York account advertising to California has schedules firing three hours “early” in local terms; a national campaign smears every local pattern across the grid (7 p.m. in the report is dinner time on one coast and commute time on the other). The blunter version of the mistake: analyzing your heatmap in your local time while the account sits in another timezone, then scheduling around a pattern that’s shifted by hours.

Setting bid adjustments on Smart Bidding and believing they work. Covered above, but it earns a second mention because the interface will not stop you. If the campaign runs tCPA/tROAS/Max Conversions/Max Conversion Value, schedule bid adjustments are decoration.

Copying someone else’s “best hours.” Templates like “pause overnight, boost 9-to-5” encode someone else’s business. Plenty of accounts make their cheapest conversions overnight, when competitors who read the same article have all switched off and CPCs drop. Your grid or nothing.

FAQ

Does dayparting work with Smart Bidding?

Partially. The schedule (on/off windows) is always honored, on every strategy. Schedule bid adjustments are ignored by Target CPA, Target ROAS, Maximize Conversions, and Maximize Conversion Value — Smart Bidding already factors time of day into per-auction bids and disregards your manual hourly modifiers. Under Smart Bidding, dayparting means: on/off windows for hours you genuinely can’t or won’t serve, campaign splits when a daypart needs its own target and budget, and hourly analysis to audit what the algorithm is doing with your money.

What’s the best time of day to run Google Ads?

There is no universal answer, and any article that gives you one is guessing. “Best hours” depend on your market, your customers, your competitors’ schedules, and what a conversion is worth to you — and they drift over time. B2B accounts often skew toward weekday business hours and e-commerce toward evenings, but those are hypotheses to test against your own 7×24 data, not rules. Sometimes the “worst” hours are the most profitable ones, precisely because everyone else switched off.

How much data do I need before changing my schedule?

Per cell you’re judging: roughly 30+ clicks and $25+ of spend before a zero-conversion cell means anything, and 3+ conversions before comparing a cell’s CPA/ROAS to the campaign average. In calendar terms that’s usually at least 4–8 weeks for a mid-sized campaign — and longer if you’re slicing to a single campaign rather than the account. Exclude the trailing few days from conversion judgments (conversion lag), and compare against a previous period to rule out seasonality.

Does Performance Max support dayparting?

Effectively no. As of mid-2026, Performance Max doesn’t give you hour-of-day bid controls, and scheduling options are minimal to nonexistent (narrow exceptions exist around store-hours for local goals). PMax decides when to serve. If time-of-day control matters for a product line, that traffic needs to live in a Search campaign where schedules apply — one more entry on the list of controls you trade away with PMax, alongside search-term visibility.

Do ad schedules use my timezone or my customers’?

Neither, necessarily: they use your Google Ads account’s timezone, set when the account was created. If your customers are in other timezones, you have to do the offset math yourself — and a multi-timezone audience means one schedule can’t be “correct” for everyone; campaign-splitting by geography is the workaround when it matters.

See your own grid

Every claim in this article is checkable against your own account in about two minutes: run the free Google Ads audit, open the Dayparting Heatmap, and look at where your week actually spends and converts. If you’re comparing tools for this kind of analysis more broadly, our honest rundown of the best Google Ads management software in 2026 covers where each one is strong — including where we’re not.

Dayparting in 2026 isn’t the bid-adjustment ritual the old guides describe. It’s knowing your 168 cells, trusting only the ones with enough data, and using the two levers that still actually move: where the schedule allows serving, and where the money goes.