What Dayparting Actually Is
At 10 AM, a shopper views your product, clicks, and buys. At 10 PM, another shopper clicks but browses competitors and abandons. Same keyword. Same bid. Completely different outcomes.
This parallels retail staffing logic — you wouldn't staff the same number of people at 3 AM as Saturday noon. The same principle applies to ad bidding.
Hourly + Weekly: Both Matter
Time-of-Day Patterns
- Increase bids 8 AM-12 PM during peak CVR
- Decrease bids 10 PM-6 AM during browsing hours
- Daytime converts ~20-30% better than evening
Day-of-Week Patterns
- B2B: strong Mon-Fri, weak weekends
- Consumer: often consistent, Sunday evening peak
- Impulse: weekends can outperform weekdays
Dayparting is an advanced refinement layer, not foundational infrastructure. If your bidding fundamentals, targeting, and campaign architecture aren't solid, fix those first. Dayparting on a broken account is rearranging deck chairs.
Choosing the Right Tool
| Option | Best For | Limitation |
|---|---|---|
| Amazon Native Rules | Small accounts (<10 campaigns) | Campaign-by-campaign setup, no bulk, no visual builder |
| PPC Platform | Accounts at scale (50+ campaigns) | Monthly cost, learning curve |
| Manual Bid Changes | Never | Requires constant monitoring, clutters bid history |
The best approach uses percentage modifiers applied on top of base bids — not changes to the base bid itself. Your base bid stays clean, modifiers layer on top, and optimization history remains readable. Whether native rules or platform-managed, the base bid should always represent the true base bid.
Pulling Your Hourly Data
Download Campaign Report with Hourly Breakdown
Amazon Ads console → Reports → SP Campaign report. Select 60-90 day range with hourly time unit. More data = more accurate conclusions.
Add a Weekday Column
Reports include date but no day-of-week. Add =WEEKDAY() formula. Essential for weekparting analysis. Note: Amazon reports default to Pacific Time.
Build Two Pivot Tables
Hourly pivot (rows = hour, values = impressions/clicks/spend/orders/sales). Weekly pivot (rows = weekday). Copy as values, then calculate metrics manually.
Calculate Key Metrics
CPC = Spend ÷ Clicks. CVR = Orders ÷ Clicks. ACOS = Spend ÷ Sales. RPC = Sales ÷ Clicks. Never drag ACOS into pivot table value field.
Dragging "ACOS" into a pivot table value field defaults to SUM or AVERAGE of individual row ACOS values — this is mathematically wrong. You can't sum or average percentages that way. Always paste pivot output as values and calculate ACOS from aggregated totals (total spend ÷ total sales).
Finding the Conversion Trends
The #1 Dayparting Mistake
Chase Order Volume
- "Most orders happen 6-10 PM — bid up evenings!"
- More evening orders exist because more traffic exists
- Higher traffic with lower CVR = higher cost-per-sale
- Paying more per click when each click is least likely to convert
Chase Conversion Rate
- Look at CVR and RPC by hour — when does each click best convert?
- High-CVR hours get bid increases even with lower volume
- Volume follows bids — you'll win more auctions in peak hours
- Result: 22% sales lift at virtually identical ACOS
Build a Conversion Heatmap
| Time Block (PT) | CVR | RPC | ACOS | Spend Share | Signal |
|---|---|---|---|---|---|
| 12-4 AM | 5.8% | $1.20 | 68% | 4% | Low volume, low CVR |
| 5-7 AM | 14.2% | $5.10 | 28% | 6% | Hidden gem |
| 8 AM-12 PM | 16.8% | $6.40 | 24% | 22% | Peak zone — bid up |
| 1-4 PM | 12.1% | $4.30 | 33% | 20% | Solid — keep baseline |
| 5-8 PM | 9.4% | $3.10 | 44% | 28% | High volume, low CVR |
| 9-11 PM | 7.2% | $2.00 | 52% | 20% | Browsing — bid down |
If total RPC is $4.00 and the 9 AM hour shows $5.40, deviation is +35%. If 9 PM shows $2.40, deviation is -40%. These deviations become the basis for your bid adjustments.
Calculating Bid Adjustments
Amazon's scheduled rules only allow bid increases. You can't directly decrease bids via rule. This requires a specific approach:
- Set base bids 10-15% below your current optimal — this becomes the "off-peak" rate
- Add scheduled increase during peak hours — typically +25% from 8 AM-2 PM
- Rest of day runs at the lowered base — effectively acting as a decrease for non-peak
If your peak hours show 25%+ better conversion rates than average, start with a 25% bid increase during those hours and a 10-15% base bid reduction. Monitor for two weeks before expanding.
The Compounding Trap
Multiple simultaneous rules multiply, not add. And placement adjustments compound on top of dayparting rules.
A "modest" 25% day rule + 25% hour rule + 100% ToS turns your $1.00 bid into $3.12. That's 3x the base bid. Calculate the compounding math before deploying.
Want Dayparting Done Right for Your Account?
I'll pull your hourly data, build the heatmap, calculate the adjustments, and deploy — no guesswork.
Three Ways to Ruin Dayparting
Following Hourly Sales Volume
90% of sellers make this mistake. Bidding up when most orders happen means paying more during hours with the worst cost-per-sale. Chase CVR, not volume.
Daypausing
Pausing campaigns entirely during "bad" hours. People still buy at midnight — CVR is lower, not zero. Lower bids instead. Let competitors overpay while you capture cheap conversions.
"Day Budgeting"
Adjusting daily budgets by hour. Amazon budgets are binary (in or out). Budget changes can't modulate CPC — the entire lever that matters. Worst of both worlds.
Daypausing vs. Bid Adjustment
All-or-Nothing
- Like negating a high-ACOS keyword instead of lowering its bid
- You forfeit ALL potential sales from that period
- Competitors get every sale you abandoned
- Midnight-4 AM is ~1% of daily budget anyway — meaningless savings
Proportional Control
- Lower bid until CPC reflects lower conversion rate
- Still capture sales at profitable CPCs
- Competitors overpay while you get cheap conversions
- Total spend stays similar — just redistributed to peak hours
Implementing & Iterating
What to Expect After Deployment
| Metric | Expected Change | Why |
|---|---|---|
| Total ad spend | Roughly the same | You're redistributing, not adding or cutting |
| ACOS | Stabilizes | Big swings between peak/off-peak smooth out |
| Total sales | Increases | More clicks during high-CVR windows |
| Peak hour CPCs | Increase (by design) | You're bidding more aggressively in buying windows |
| Off-peak CPCs | Decrease | Lowered base bids reduce off-peak spend |
Allow at least two weeks before evaluating results. You're changing the CPC curve, which changes spend distribution, which changes where impressions and clicks land. Give it time to stabilize.
When Dayparting Isn't Worth It
Small accounts with low volume ($30/day)
Hourly trends lack data reliability. Focus on bidding and targeting fundamentals first. You need enough volume per hour for the patterns to be statistically meaningful.
Flat intraday curves
If conversion rate barely moves between 8 AM and 10 PM, there's nothing to optimize around. Check your data first — don't assume dayparting is needed.
Accounts with bigger problems
ACOS at 80% and campaign structure is messy? Dayparting won't save you. Fix the foundation first — bidding, targeting, structure — then come back to time-based optimization.
Pull your hourly data and check RPC deviation from average. Seeing 20%+ swings between best and worst hours? Dayparting will make a real difference. Under 10% deviation? Probably not worth the effort.
Volume follows bids. You don't just passively respond to the hourly sales curve — you reshape it. Accounts where evening hours dominated orders now show inverted curves after dayparting, with more daytime sales, because they became more competitive in morning auctions.— Mohsin Raza, Adaptoid E-Commerce
- Chase conversion rate, not order volume. Bid up during hours with the highest CVR and RPC — not the hours with the most total orders. Volume follows bids.
- Use percentage modifiers, not base bid changes. Keep your base bid clean. Layer time-based adjustments on top so optimization history stays readable.
- Start conservative: +25% peak, -15% base. Two simultaneous 25% rules compound to 56%. With ToS placement, a $1 bid becomes $3.12. Calculate the math before deploying.
- Never daypause. Lower conversion rates need lower bids, not complete shutdown. You still capture sales at CPCs that reflect the lower CVR.
- Never "day budget." Amazon budgets are binary (in or out). Budget changes can't modulate CPC. Use bids, period.
- Wait two weeks before judging. Dayparting changes the CPC curve, spend distribution, and click patterns. Give it time to stabilize before evaluating.
- Cross-reference hours with days. "Wednesday mornings" might be the goldmine — not "Wednesdays" or "mornings" generally. Filter hourly data by weekday for the sharpest insights.
- Apply the 20% litmus test. If your best-to-worst hourly RPC deviation is under 10%, dayparting probably isn't worth the effort. Over 20%? Absolutely deploy it.