The Amazon Ads
Dayparting Starter Guide.

Most Amazon advertisers spend more time optimizing their bids than they spend thinking about when their ads run. That is the wrong priority. A well-timed $1.50 bid outperforms a poorly-timed $3.00 bid every time — because the shopper on the other end of the well-timed one is actually ready to buy.

Dayparting is the practice of controlling when your campaigns run. It is not about bidding less. It is about making sure the budget you do spend lands in the windows where your customers convert, not the windows where they browse, compare, and close the tab. This guide gives you the framework to build your first schedule, the category-specific benchmarks to start from, and the five-step setup to get it live.

Dayparting is scheduling. Not bid strategy.

Dayparting gets grouped with bid optimization in most guides, which is misleading. Bids control how much you pay per click. Dayparting controls whether your campaigns are even eligible to serve during a given hour. They are different levers with different effects — and dayparting is often the higher-leverage one because it operates at the campaign level rather than the keyword level.

In practice, dayparting for Amazon Sponsored Products means one of three things. You can pause campaigns entirely during low-performing hours and resume them automatically. You can reduce bids during hours where conversion rate is low — staying in the auction but at a lower price. Or you can increase bids during your peak conversion windows to capture more of the demand while it is there. A complete dayparting strategy uses all three, applied to different campaign types based on their data.

The time-of-day pattern matters because Amazon shoppers are not uniformly active across 24 hours. They browse during the day, decide in the evening, and buy in a predictable window that varies by category. When your ads run in windows where nobody is converting, you are not just wasting that budget — you are also diluting your campaign's overall conversion rate signal, which matters for keyword ranking. Running fewer, better-timed impressions can improve both efficiency and organic rank.

Here is what this looks like in practice. A seller running $150/day across all 24 hours might see 3 AM–6 AM generating $8 in spend and zero orders, while 7 PM–9 PM generates $12 in spend and 6 orders. The ratio metrics — ACoS, conversion rate — look terrible overnight and excellent in the evening. But the 7-day average blends both windows together and the account shows a "reasonable" 32% ACoS. A dayparting rule redirects the $8 from 3 AM into 7 PM. Same total spend. Two extra orders. 32% ACoS becomes 24% ACoS. Nothing else changed.

The 7-day conversion heatmap.

The heatmap below shows general conversion patterns across the US Amazon marketplace, based on Perpetua's 2022 intraday study — the most rigorous published research on hourly Sponsored Products performance. These are directional starting points. Your category, price point, and customer base will all shift the pattern. Use this as a first hypothesis and validate it against your own Hourly SP Report data.

Two patterns hold consistently across the Perpetua data. Weekday evenings — 6 to 9 PM Pacific — are where conversion rates are highest and ACoS runs lowest. The overnight window from midnight to roughly 5 AM is where ACoS is highest and order volume is smallest. Saturday morning is the strongest single window of the week for most general merchandise, home, and outdoor categories.

Best hours by category.

The heatmap above is cross-category. In practice, different product types have meaningfully different peak windows based on who buys them and when they make purchase decisions. The table below gives starting points for the most common Amazon categories. These are directional — validate them against your own hourly data before making them permanent.

Use your category row as your starting hypothesis. If your product sits at the edge of multiple categories — a fitness gadget, a kitchen organization product — start with whichever pattern fits the purchase occasion, not just the browse occasion. A vitamins brand targets the evening replenishment window, not the midday browse window, even though health products get browsed all day.

Start with one pause rule. Midnight to 6 AM.

Every account is different, but the midnight-to-6-AM overnight pause is the highest-impact first rule for the vast majority of Amazon advertisers. Here is why it works so reliably. Conversion rates in this window are structurally low across nearly every category because the population of shoppers active at 3 AM is tiny and not in buying mode. At the same time, CPCs in this window can remain elevated because automated bidding tools continue to compete. The result is spend that generates almost no orders.

Pausing overnight does not mean your total impressions drop by 25%. Overnight hours account for a small fraction of total daily conversions — the Perpetua data shows this window generating roughly 5 to 10% of daily orders for most accounts while consuming 20 to 30% of daily budget at elevated ACoS. Reallocating that budget to your 6–9 PM window means the same total spend drives more total orders.

One important caveat: if your product is in the first 60 days of launch, hold off on full overnight pauses. During launch, your campaigns are partially a ranking investment — even low-volume overnight conversions contribute to the sales velocity and keyword relevance signals Amazon's algorithm uses to rank your product. In launch phase, use a bid reduction (20 to 30% lower overnight) rather than a full pause. Once you are past 60 days with stable ranking, convert that to a pause if the data supports it.

Build your first schedule.

Set it up in 5 minutes.

Once you know what schedule you want to run, here is the exact sequence to get it live. This assumes you are setting up your first Dayparting Rule in Off Hours. The whole process takes under five minutes for a single rule.

How to know if your schedule is working.

Two weeks after activating your first dayparting rule, pull the Hourly SP Report again and compare the same hours before and after. This is the only reliable way to evaluate a schedule — the overall account metrics lag by one to two weeks and can mask what is actually happening at the hour level.

Here is what a working schedule looks like in the data. Your paused hours should show zero or near-zero spend. Your peak hours should show higher spend than before — that is the budget shifting. And your ACoS in peak hours should hold steady or improve as the concentrated impressions deliver better conversion rates. If you see peak hour ACoS rising alongside spend, that usually means competition in that window is high and a bid increase is needed to win share. If you see peak hour spend holding flat despite the rule, your daily budget cap may still be too low to absorb the redirect.

What a rule that is not working looks like: your paused hours still show spend. This is almost always a rule configuration issue — the hours in Off Hours may be set to the wrong timezone, or the rule may not be assigned to the right campaigns. Check the Off Hours change log, which shows every rule fire with a timestamp, and confirm the campaigns in the pause window match the ones generating that overnight spend in the report.

What to track week over week

The longer-term signal to watch is TACoS — Total ACoS, which divides total ad spend by total revenue including organic. As paid efficiency improves, your conversion rate signal strengthens, organic rank gradually improves, and organic sales grow. When organic sales grow, TACoS falls even if ACoS stays flat. Give it 4 to 6 weeks for TACoS to reflect the effect. If TACoS is flat or rising after 6 weeks, something in the strategy needs review — most likely the pause window is too aggressive or a product is in earlier stages of its lifecycle than expected.

When to hold off — and when to be careful.

Launch phase (first 60 days). If you are still building keyword rank and review count, overnight conversions — even expensive ones — contribute to the sales velocity signals Amazon's algorithm uses for ranking. Full pauses during launch can slow that momentum. Use a bid reduction (20–30%) instead of a pause until you are past the launch window.

Thin data. If a campaign has fewer than 14 days of history or your hour-level order counts are in the single digits, you are looking at noise rather than patterns. Wait for the data to thicken before building rules around it. Acting on thin data is how sellers build schedules that hurt more than they help.

Highly seasonal categories. Your August data looks nothing like your October data if you sell Halloween costumes. Rules built on off-peak patterns will be wrong during peak seasons. Revisit your hourly analysis at the start of each major selling window and rebuild rules accordingly.

Starting too complex. It is tempting to build a custom rule for every hour of every day once you have the data. Resist this. Start with one pause rule and one boost rule. Add complexity only when you have data showing the addition will improve performance. A two-rule schedule you understand beats a twelve-rule schedule you cannot diagnose when something goes wrong.