There is a specific kind of ad spend that is almost invisible. It does not show up as a crisis on any dashboard. It is not tied to a broken campaign or a rogue keyword. It is quietly happening while everyone is asleep. And if you have not specifically looked for it, you are probably doing it.
Overnight Amazon ad spend. The 2 AM impressions, the 3:30 AM click, the 4 AM budget draw-down that happens in accounts that are set to run 24 hours a day because nobody ever told them not to.
In isolation, this looks harmless. A few dollars an hour. Occasional conversions. No obvious damage. But the shape of the waste is what makes it expensive, and the compounding effect is what makes it matter. Here is how to see it in your own account, what to do about it, and why the amount of money at stake is bigger than the per-night dollar figure suggests.
The premise
Every hour your ads run, three things are happening.
One, your daily budget is being consumed. Amazon does not pace budget perfectly across 24 hours, but it does allocate a portion of the budget to each hour based on impression availability and competition. If you have a $200 daily budget and you are running 24 hours, some non-trivial share of that budget is being spent in hours when almost nobody is converting.
Two, you are paying the market rate for clicks during those hours. CPCs overnight are often only marginally lower than daytime CPCs in most categories. The bid landscape does not collapse at midnight. You are paying roughly the same price per click, for clicks that convert at a fraction of the daytime rate.
Three, you are diluting your account-level conversion rate. Amazon's organic ranking is influenced by sales velocity and conversion rate. Low-intent traffic drives down the conversion rate signal without meaningfully increasing velocity. The effect is small but real.
Stack all three together, across every account in a portfolio, across every night of the year, and the number stops being small.
A representative scenario
Consider a hypothetical home goods brand. Standard Sponsored Products setup, running across a catalog of maybe 40 ASINs, typical daily ad budget in the low thousands across all campaigns. Ads are set to run 24 hours a day. This is the default, and this is what most accounts look like when no one has specifically done anything about it.
If you pulled a representative week of hourly data from an account like this and looked at the 12 AM to 5 AM window, you would typically see a pattern that looks like this:
- Impression volume is roughly 20 to 35 percent of daytime levels
- CPC is usually within 10 percent of daytime CPC, sometimes slightly lower, sometimes not
- Click-through rate is similar or slightly lower
- Conversion rate is 30 to 60 percent of the daytime rate
- ACOS during this window is often 2x to 4x the daytime ACOS
The specific numbers vary by category and account. The shape is remarkably consistent.
Put differently. Overnight ad dollars, in most categories, buy you roughly a third of the performance per dollar that your daytime ad dollars buy you. You are paying a premium for underperforming traffic.
Why this goes unnoticed
Three reasons, mostly.
First, it looks small in absolute dollars. Five hours a night at a fraction of normal impression volume does not generate a big line item in any dashboard. You might be talking about $15 to $40 per day of ad spend, per account, going to this window. In isolation, that reads as noise.
Second, there are still conversions in that window. Not zero. Someone is always buying at 3 AM. When you spot-check the data, you see sales attributed to overnight clicks and assume the hour is pulling its weight. What you are not comparing is what that same dollar would have done in a daytime hour.
Third, most dashboards aggregate by day, not by hour. The standard Amazon Advertising reports default to daily totals. You have to deliberately pull hourly data to see the pattern. Most accounts never do.
The combination of those three is why accounts run this way for years without anyone noticing.
The math that makes this matter
The small dollars compound in ways that are less obvious.
Take the home goods scenario above. Suppose the account is spending, illustratively, $30 a night on ads between midnight and 5 AM. That is roughly $900 a month, or $10,800 a year, per brand.
Run that scenario across an agency managing ten brands, and the line item is $108,000 a year in ad spend that is producing about a third of the return per dollar that daytime spend is producing. Not all of that money is wasted. Some of it is buying real conversions. But if you reallocated even half of it to high-intent daytime windows where conversion rates are three times higher, the effective return on that reallocated budget increases meaningfully.
This is the compounding piece. Overnight spend is rarely a crisis for any single account, but it is a structural inefficiency that scales with the size of the portfolio. The agency managing 50 accounts is walking past a five-figure annual opportunity in most of them.
What to actually do about it
Step one is to measure what is happening in your own account. Do not take the scenario above as gospel. Your numbers will be different. The process:
- Pull 30 days of hourly performance data from Amazon Advertising reports
- Filter to the overnight window (midnight to 5 AM, or whatever you want to define)
- Compare impressions, clicks, spend, orders, sales, ACOS, and conversion rate for that window against your daytime baseline
- Calculate the spend-to-sales ratio for that window specifically
- Ask: if you reallocated that budget to peak hours, what would the expected return be?
Most accounts will come out of this exercise with a clear directional answer. Overnight is underperforming. The question is only by how much.
Step two is to decide how aggressive to be. The options, in rough order of conservatism:
Conservative. Pause ads midnight to 5 AM on weeknights only. Five hours per night, five nights per week. 25 hours paused out of 168. Minimal risk to weekend peaks or early morning buyers.
Moderate. Pause midnight to 6 AM every night. Roughly 42 hours paused per week. Covers the full dead zone in most categories.
Aggressive. Pause midnight to 6 AM every night plus a post-peak wind-down from 11 PM to midnight. Roughly 49 hours paused per week. Appropriate for categories with particularly weak late-night performance (kitchen, home goods, office supplies).
Step three is to actually implement it. This is where the problem usually dies. The obvious answer is to log into Amazon every night at midnight and toggle campaigns off, then log in again at 6 AM and toggle them on. Nobody does this. It is not realistic.
The realistic answers are:
- Set up dayparting rules in whatever automation tool you use
- Build the schedule once, let the tool execute it 168 times a week
- Review quarterly, not daily
Off Hours exists for exactly this problem. Schedule your overnight pauses once, the system handles the toggling, you never think about it again. Other tools (Helium 10 Adtomic, rule-based platforms from the suite players) can execute dayparting as well. What matters is that some tool is doing it.
Stop paying for 2 AM clicks.
Off Hours executes your overnight pause schedule 168 times a week. Set it once. 14 days free.
Start free trial →The three questions to ask your account
If you want a quick diagnostic without pulling a full report, these are the three questions that will tell you most of what you need to know.
Question one: what is your conversion rate between midnight and 5 AM, compared to your daily average?
If it is meaningfully lower (under 70 percent of daily average), you are likely overspending during those hours. If it is similar or higher (rare, but exists in a few categories), leave it alone.
Question two: what share of your daily budget is being consumed overnight?
Pull the hourly spend report, add up the spend between midnight and 5 AM, divide by total daily spend. If overnight is taking more than 15 percent of daily budget in most categories, that budget is probably doing less work than it could.
Question three: how often does your daily budget exhaust before the evening peak?
If your ads regularly run out of budget before 6 PM, overnight spend is costing you twice. Once in the form of lower-converting traffic, and again in the form of high-intent traffic at 7 PM that you cannot serve because the budget is already gone.
If the answers to any of these three are bad, dayparting is worth doing. If the answers are all clean, you may be in one of the small number of categories (insomniac-friendly niches, certain B2B, global brands with non-US traffic) where 24-hour running actually works. Most accounts are not in those categories.
The second-order effects
A few things happen when you stop burning budget overnight that do not show up in the obvious places.
Your daytime CPC can sometimes drop. This sounds counterintuitive. The mechanism is that when your daily budget is no longer being depleted overnight, you are not forced to drop bids to stretch budget across the day. Cleaner budget availability during peak hours often translates into more consistent auction participation.
Your account-level conversion rate improves. Pulling low-converting impressions out of the account raises the aggregate number. This has small but real implications for Amazon's organic ranking signals.
Your ACOS reports get cleaner. Without the drag of overnight hours, the daily ACOS number more accurately reflects your actual campaign performance. Decisions made based on ACOS become more reliable.
Your team stops spot-checking the account at weird hours. This is the soft benefit nobody mentions. Media buyers who know the account runs 24/7 sometimes feel compelled to check performance in the evening or early morning. When the schedule is explicit, there is less to check.
What to do next
If you take one thing from this post, let it be: pull your hourly data before the end of the week. Do not take the scenario numbers above as your account's numbers. Your account will tell you its own story, and the answer is almost always smaller or larger than you expected.
If the data tells you overnight is underperforming, build a dayparting schedule. Conservative first. Tighten later.
For the complete guide to fixing ACoS driven by overnight spend, read how to reduce Amazon ACoS.