This is the deep-dive reference for Step 10 of optimization-weekly.md. The top-level routine has the budget-pacing tune in checklist form. Open this doc when you need: the math worked example, the why behind narrowing schedules, the Maximize Clicks vs Smart Bidding mechanics, or guidance on edge cases.
When to use this doc
- Client has a small budget (under $50/day) and CPCs are eating it up
- CPL is drifting up week-over-week and you suspect uneven daily/hourly performance
- Specific days (often weekends) are consistently underperforming
- The budget hits its daily cap before end-of-day on most days
- A new operator is learning when narrowing the schedule is the right move vs raising the budget
When this matters most
the highest-leverage scenario is **clients with budgets under 30 click can blow the daily budget and the campaign goes silent for the rest of the day.
Higher-budget campaigns (>$100/day) benefit from ad schedule tuning too, but the leverage is smaller. The tactic is still valid; the urgency is lower.
The core insight
If your client has a $1,000/month budget and you run 24/7 across all days, the math is:
- Daily: $33
- With 30+ days
If you switch to Mon-Fri only (same $1,000 monthly):
- Daily (Mon-Fri only): ~$48
- With $14 CPC: 3-4 clicks per day on weekdays
You don’t ask for more money; you reallocate what they have to higher-performing windows. Same total monthly cost, ~50% more clicks where conversions actually happen.
The decision tree
Before you narrow anything, run through this tree:
Question 1: Is CPL actually drifting up week-over-week?
If CPL is stable, ad schedule tuning isn’t urgent. Skip this routine for the week and revisit next week. Don’t fix what isn’t broken.
If CPL is drifting up but you’ve already paused bleeders (Step 2 of weekly) and added negatives (Step 3), and the CPA drift checks (Steps 4-5) didn’t surface a cause, ad schedule pacing is the next likely lever.
Question 2: Does the campaign hit its daily cap before end-of-day on most days?
Look at the Average daily spend over the last 14 days. If it’s hitting 33 most days, the budget is exhausted before the day ends. Two paths from here:
- Path A (default): narrow the schedule to concentrate budget on better windows
- Path B: ask the client for a budget increase (always preferable, but often refused)
Question 3: Are specific days/hours consistently underperforming?
Open Reports → Day and Hour breakdowns over the last 14-30 days. Look for:
- Days with lower conversion rate (consistently across multiple weeks, not just one anomaly)
- Hours with high cost and low conversion rate (e.g., 3 AM HVAC searches that don’t convert)
If you can identify clear patterns, you have actionable data for narrowing.
Question 4: Bidding state, Maximize Clicks or Smart Bidding?
The mechanics differ. Apply the right tactic for the bidding state:
| Bidding state | What works |
|---|---|
| Maximize Clicks (or Manual CPC) | Percentage bid adjustments AND hour/day removal both work |
| Smart Bidding (Target CPA / Maximize Conversions) | Only hour/day removal works. Percentage adjustments are ignored. |
Under Smart Bidding, the only schedule lever is the hard exclusion (remove the hour or day from the schedule entirely). See for the Smart Bidding-specific treatment.
Tactic A: Narrow the schedule (remove dead days/hours)
Use this when specific days or hours consistently underperform.
Procedure
- Open campaign Settings → Ad Schedule
- Identify the days/hours to remove (from your Question 3 analysis above)
- Remove them by clicking into the schedule and deleting those rows
- Save
Worked example from A junk-removal client with a 1,000/month) budget in a $14-32 CPC market:
Before:
- Schedule: 24/7
- Daily budget: $33
- 30 days
Operator data review:
- Sundays consistently underperform (low search volume, no conversions in last 4 weeks)
- Mon-Sat conversion rate is 4x Sunday’s
After:
- Schedule: Monday-Saturday only (Sunday removed)
- Daily budget: $33 (unchanged)
- Effective spread: ~38/day** on the days that actually convert
- More clicks per active day, more impressions on high-performing days
The budget didn’t change. The client didn’t pay more. Conversions improved.
What about removing specific hours?
Same procedure as removing days, but at the hour level. Common removals:
- HVAC: remove 1 AM - 6 AM (no one’s converting at 3 AM unless emergency, and emergency searches are usually a small fraction of HVAC’s daily volume)
- Plumbing: keep 24/7 if the client takes after-hours emergency calls; otherwise remove evening hours after the client stops answering
- Roofing: remove evenings/weekends if client only does business hours
- Junk Removal: remove hours when the client can’t physically schedule pickup that day
The general rule: only show ads in hours when someone at the client can answer the phone or respond to a form within an hour or two. Otherwise the click goes to waste.
Tactic B: Apply percentage bid adjustments (Maximize Clicks only)
Use this when the data shows certain days/hours over-perform but you don’t want to fully exclude others.
Procedure (Maximize Clicks only)
- Open campaign Settings → Ad Schedule
- For each day/hour row, click the bid adjustment column
- Enter a percentage (e.g., +20% on top-converting hours, -20% on dead hours)
- Save
Calibration ranges
- +10% to +30% on top-converting hours/days
- -10% to -30% on weak hours/days
- -50% to -90% on truly dead hours (only if you don’t want to remove them entirely)
Don’t apply +50% or +100% adjustments unless the data is overwhelming. Big swings make the bid auction less predictable.
Why this doesn’t work on Smart Bidding
under Maximize Conversions or Target CPA, Google’s algorithm sets bids based on conversion likelihood, not your manual adjustment. Adding a +20% bid adjustment changes nothing. The Target CPA stays the same; the algorithm bids what it bids.
The only schedule lever that works under Smart Bidding: remove the hour or day entirely. This is a hard exclusion and the algorithm respects it.
Tactic C: Set or tighten the Max CPC cap (companion lever)
Sometimes the schedule isn’t the right lever, the cap is. If $30 single-click outliers are blowing the budget, narrowing the schedule helps a bit, but capping the cost-per-click prevents the spike entirely.
For Max CPC cap setup, see the companion deep-dive: max-cpc-cap-validation.md.
The two tactics often work together:
- Narrow schedule (remove low-converting hours) → concentrates budget on good windows
- Set Max CPC cap → prevents single-click budget burn within those windows
For an undersized-budget campaign in a high-CPC market, run both in tandem.
The caveat
narrowing the schedule is a band-aid for an undersized budget. The ideal solution is to push for a budget increase. If the client refuses, schedule narrowing makes the most of what they have. But:
- A 25 average CPC) is structurally too small. No amount of scheduling magic produces enough clicks to generate consistent conversions.
- A $200/day plumbing campaign in the same market doesn’t need this routine; budget isn’t the constraint.
Budget conversation thresholds (rough agency calibration):
- Under $40/day: budget is the constraint; schedule narrowing is a stop-gap; push hard for increase
- $40-100/day: schedule narrowing is meaningful leverage; can produce real gains without budget conversation
- Over $100/day: schedule narrowing is fine-tuning, not rescue
Always make the case for budget increase first. If the client says no, then narrow.
Edge cases
Edge case 1: client says “I want 24/7 because emergencies”
Some niches genuinely need 24/7 (emergency plumbing, water-damage restoration, towing, locksmith). For those, don’t narrow the schedule. Instead, use:
- Tighten the Max CPC cap (companion deep-dive: max-cpc-cap-validation.md)
- Add ad-group-level negatives for hours where searches don’t convert (per optimization-daily.md)
- Make the case for a budget increase (the most common right answer)
Edge case 2: schedule data is noisy (campaign is new)
Before narrowing, you need at least 2-3 weeks of day/hour data. Below that threshold, the patterns you see are noise. Wait one more weekly cycle before deciding.
Edge case 3: seasonal client with weekend demand spikes
Roofing in early autumn (storm-damage spikes), HVAC in heat waves (weekend AC failures). For seasonal clients, the historical day/hour pattern doesn’t predict the current pattern. Use the most recent 2-3 weeks of data, not all-time.
Edge case 4: undersized budget + high CPC + Smart Bidding
This is the worst-case combination. Smart Bidding ignores percentage adjustments. The only levers are:
- Remove dead hours entirely (allowed)
- Lower Target CPA (the algorithm tries harder to find cheap conversions)
- Push for budget increase
If the client refuses both budget increase and Target CPA reduction, the campaign may genuinely not be viable on Smart Bidding. Consider reverting to Maximize Clicks (per optimization-the-switch.md) where you have more levers.
Logging
After ad schedule changes, log in clients/<client>/optimization-log.md:
2026-05-06 weekly | ad schedule analysis: campaign hitting daily cap by 4pm on most days
2026-05-06 weekly | removed Sunday from schedule | reason: 0 conversions in last 4 Sundays, $40+ wasted spend
2026-05-06 weekly | removed 2-6 AM hours weekdays | reason: zero conversions, $25 wasted in last 14 days
2026-05-06 weekly | applied +20% bid adj on Tue-Thu 11am-1pm (Maximize Clicks campaign) | top-converting window
For Smart Bidding campaigns, only “removed” entries are valid (no percentage adjustments).
Back to: optimization-weekly.md