Dayparting is one of those tactics that sounds obviously correct. Spend more when shoppers buy, spend less when they don't, and watch your efficiency climb. Every PPC tool now sells an hour-of-day schedule as a premium feature, and plenty of brand owners turn one on because it feels like the sophisticated move.
Then sales drop on Tuesday afternoon and nobody can say why.
The honest answer is that dayparting helps a minority of accounts and hurts a meaningful number of the ones that adopt it carelessly. The tactic is real. The problem is that most people apply it before they have the data to justify it, and they read the data they do have backward. Let's fix both.
What dayparting actually does
Dayparting means changing your bids, or pausing campaigns entirely, based on the hour of the day or the day of the week. The logic is that conversion rate is not flat across a 24-hour cycle. If shoppers convert at 9 percent during evening hours and 4 percent at 3 a.m., a dollar of ad spend buys very different results depending on when it lands.
So far so good. The catch is that Amazon does not bill you by conversion rate. It bills you per click, and it serves your ad based on a bid you set. Dayparting only works if you can reliably separate the hours where your money works hard from the hours where it leaks, and if that pattern is stable enough to schedule against. For a lot of accounts, it is neither.
There is also a structural cost people forget. When you pause or slash bids during "bad" hours, you surrender impression share and ranking signals during that window. Amazon's algorithm rewards consistent sales velocity. Going dark for six hours a night can cost you organic position in a way that does not show up in your ad report at all. That hidden cost is exactly the kind of thing we watch for when the goal is to lower ACoS without killing your sales, because a cheaper ACoS that drags down organic rank is not a win.
When it is worth it, and when it is not
Dayparting earns its keep under a few specific conditions.
It helps when
- Your budget caps out before the day ends. If your campaigns run dry by 2 p.m., you are already dayparting by accident, just badly. Concentrating that limited budget into your strongest hours is a real gain.
- You have a genuine, repeatable conversion pattern. A B2B product that sells during office hours, or a treat product that moves at night, can show a swing large enough and consistent enough to schedule against.
- You run high volume. You need enough clicks per hour for the numbers to mean something. A campaign with three clicks in a given hour is noise, not signal.
It rarely helps when
- Your daily budget never runs out. If Amazon already serves your ad whenever a shopper searches, throttling "weak" hours just hands those sales to a competitor. You are not saving money, you are donating placement.
- Your conversion swings are small. A move from 7 to 6 percent is not a reason to go dark. It is within normal weekly variance.
- You are still early in a launch. New products need maximum exposure and velocity, not a schedule that starves them. During the first 30 days of an Amazon launch, consistency beats efficiency every time.
If your budget never runs out, dayparting is not saving you money. It is handing your best placements to whoever bid through the hour you went quiet.
How to read the data before you touch a single bid
This is where most brands go wrong. They look at spend by hour, see that 1 a.m. cost them forty dollars, and pause it. Spend is the wrong column to lead with.
Pull your hour-of-day report and sort by these three things in order.
1. Conversion rate by hour, not spend by hour. You are looking for hours where shoppers click and do not buy. That is wasted money. High spend on a high-converting hour is not waste, it is your engine.
2. Click volume per hour. Throw out any hour with too few clicks to trust. If an hour has fewer than roughly 20 to 30 clicks across your sample window, you cannot draw a conclusion from it. Treat it as average and move on.
3. A multi-week sample. One week is a coincidence. Look at four to eight weeks so a single odd day, a holiday, or a stockout does not masquerade as a pattern. Amazon reports time in a fixed time zone, so remember your buyers span multiple zones. The "3 a.m." dip on your screen is breakfast somewhere.
Only after those three filters do you look at cost. Now the question is sharp: are there specific hours that consistently take spend and consistently fail to convert, with enough volume to be sure? If yes, you have a dayparting candidate. If the chart is just gentle noise, you do not, and a schedule will cost you more than it saves.
This is the same discipline that makes a search term report readable like a strategist instead of a spreadsheet you stare at. Sort by the decision you are trying to make, filter out the noise, then act.
Tie the decision to margin, not ACoS
Even a clean hourly pattern does not automatically justify dayparting. The real test is contribution margin per hour, not ACoS per hour. An hour with a 40 percent ACoS can still be your most profitable window if it drives full-price sales and feeds organic rank. An hour with a 15 percent ACoS can be a loser if those orders come with heavy coupons or returns.
Before you cut any window, confirm the move protects profit, not just the dashboard. We have written about why contribution margin should drive every Amazon decision, and dayparting is a perfect example. The hour that looks expensive on an ACoS chart may be carrying your whole funnel.
There is also a placement angle. If your "weak" hours are weak only on Sponsored Products but strong on Sponsored Brands, the answer is not a schedule, it is a reallocation. Deciding where each ad dollar works hardest between Sponsored Brands and Sponsored Products often solves the problem you were about to throw a dayparting rule at.
What to do this week
Run the test before you run the tactic. Here is the order.
- Pull six to eight weeks of hour-of-day data. Use Amazon's reporting or your bid tool's hourly export.
- Sort by conversion rate, then drop low-click hours. Anything under roughly 20 clicks in the window gets marked "not enough data."
- Find consistent low-converting, high-spend hours. If three or more separate weeks point at the same window, you have a real pattern. If not, stop here and leave bids alone.
- Check the budget question. If your campaigns never run out of budget, do not daypart. Fix the budget cap or the bids instead.
- Apply a small reduction first. Lower bids by 20 to 30 percent in the proven weak hours rather than pausing outright. Watch organic rank and total units for two weeks, not just ACoS.
Dayparting is a scalpel, not a default setting. Used on the right account with clean data behind it, it trims real waste. Used on faith, it quietly hands your shelf to a competitor while your ACoS report looks great. If you want a second set of eyes on your hour-of-day numbers before you change anything, that is exactly the kind of read we do every day.