"How does Ralph solve dayparting" is a fair question and it deserves the plumbing, not adjectives. This page is the plumbing. If you want the argument for why a desk belongs next to Ads Manager at all, that is why Ralph for dayparting. This is what actually happens.
Direct answer: Ralph ingests hour-of-week performance from your Meta account, draws a seven-day heatmap, and proposes which hours to switch off or throttle, with floors so it never darkens half the week on thin data. You approve. On lifetime-budget Meta ad sets it then writes a native ad schedule. On daily-budget ad sets, which is most Shopify accounts, it runs a stepped budget throttle instead, because Meta has no clock there. On Google it recommends the schedule and you apply it. It does not auto-daypart, it does not let weather move the hours, and it does not write anything without a yes.
The problem, stated plainly
Most Shopify brands run Meta on daily budgets. Meta only offers ad scheduling on lifetime budgets. So the account has hours that clearly lose money, usually overnight, and no native switch to turn them off. People either move to lifetime budgets and accept the learning reset, write automated rules that slam budgets up and down, or give up. The why is in why Meta dayparting doesn't work.
Google is easier mechanically, since Search and standard Shopping have a proper ad schedule, but the evidence still lives in a breakdown report nobody opens, and Smart Bidding quietly ignores the bid adjustments you set.
Step one: hourly evidence, in the account's time zone
Ralph pulls hourly performance rows for the connected Meta account and stores enough of them to see a weekly pattern. That last part matters. A dayparting decision made on a week of data is a decision made on a promotion, a bank holiday or a bad creative. The heatmap is only shown when there is enough behind it, and if the overnight you already know is dead does not show up, that is a sync or time zone bug to fix before anything else.
Step two: the heatmap and the retail blocks
Seven days across, 24 hours down. Spend, results and cost per result in each cell. Rather than asking you to read 168 cells, Ralph groups them into retail blocks that match how shopping actually happens: overnight (0 to 6), breakfast (6 to 9), daytime (9 to 14), afternoon (14 to 17), evening (17 to 22) and late (22 to 24). Blocks are compared to the account average, not to a theory about when people shop.
Step three: a proposal with floors
A block is only proposed as OFF when it is confidently below the account average, not merely a bit under it. Then the rails apply:
- Never OFF more than half the week.
- Never a whole day OFF.
- No proposal at all on empty or thin data.
Those rails exist because the failure mode of every dayparting tool is over-fitting: it finds four dead cells and darkens Tuesday. The floors make the proposal boring, and boring is what you want from something that touches budget.
Step three and a half: is it the clock, or the shelf?
This is the part most tools cannot do because they only see the ad account. For a fashion or footwear brand, the evening often looks weak because the sizes that convert sold out during the day. That is not an hour-of-day effect. It is an empty shelf, and switching the evening off would hide the real problem. Ralph sits next to variant stock, so the proposal can say: the evening is weak, and here is the size run that broke at three o'clock. Fix the shelf, or exclude the item, before you touch the hours. That whole trap is written up in dayparting for fashion brands.
Step four: you approve
The proposal lands in review with the evidence attached. Approve, reject, or edit the cut. Nothing has been written to the ad account yet. On the Free plan this is where it stops: you can see the heatmap and the proposal and act on it yourself in Ads Manager. Growth and above can approve and let Ralph apply.
Step five: the write, which depends on the budget type
| Ad set | What Ralph writes after approval | What it looks like in Ads Manager |
|---|---|---|
| Meta, lifetime budget | A native ad schedule with the approved hours | The ad set's Ad scheduling setting, exactly as if you had set it |
| Meta, daily budget | A stepped budget throttle: from baseline, step down around 20% per tick in weak hours, floor around 20% of baseline, hold if the ad set is in learning, restore to baseline afterwards | Budget changes in the change history, with the reason. No schedule, because Meta has none |
| Google Search or Shopping | Nothing. A recommendation with the hours and suggested bid adjustments | You set the ad schedule in Google Ads yourself |
| Advantage+ shopping | Nothing, and the proposal will say to leave it wide | Unchanged |
The throttle deserves its own explanation because it is the bit people get wrong when they build it from automated rules: daily-budget Meta dayparting: how throttling actually works.
Three things Ralph will not do
- Auto-daypart. There is no mode where it writes schedules without a human. The writers are off by default and stay off until you enable them.
- Let weather move the hours. Weather is a demand signal in the brief. Rain at 3pm should not pause your ads at 3pm, and a product that does that should not exist. The reasoning is in weather-based dayparting vs weather-triggered ads.
- Claim a Google write it does not have. Google is recommend-only today. If that changes, this page will change.
A dayparting tool's job is to remember the clock when you are not in Ads Manager, and to refuse to invent a pattern when there is not one. Everything else is decoration.
A fair way to evaluate it
Do it yourself first. Pull the hourly breakdown in Ads Manager for sixty days, name one dead overnight in writing, check those hours still spend, check the SKUs they sell still have a size run. Then connect Meta to Ralph and open the daypart view. If the dead overnight you named is invisible, we have a bug. If the proposal wants to darken half the week, reject it and tell us. If it proposes the one cut you would have made, the desk is doing its job. The DIY steps are in how to daypart Meta ads and Google Ads dayparting.
What happens between connecting an ad account and a schedule going live
- Ingest hourly history. Ralph pulls hour-of-week performance for the ad account in the account's time zone and keeps enough history to see a pattern, not a week.
- Draw the heatmap. A seven-day by 24-hour grid of spend and results, grouped into retail blocks: overnight, breakfast, daytime, afternoon, evening, late.
- Propose a cut with floors. Only hours confidently below average are proposed as OFF. Never more than half the week, never a whole day, never on thin data. Empty data means no proposal.
- Check stock before blaming the clock. If the weak hours coincide with the converting sizes selling out, the proposal says so, because that is a shelf problem, not an hour problem.
- You approve, then it applies. Lifetime-budget Meta ad sets take a native ad schedule. Daily-budget ad sets take a stepped budget throttle. Google gets a recommendation you apply yourself. Nothing writes without a yes.
Questions people actually ask
Does Ralph automatically daypart my ads?
No. It proposes. Ralph reads hourly history, draws the heatmap, and suggests weak hours to switch off or throttle, with coverage floors. A human approves before anything is written to the ad account, and the writers are off by default.
How does Ralph daypart a Meta ad set on a daily budget?
Meta has no native schedule on daily budgets, so Ralph uses a stepped throttle: dim the budget from baseline in steps of around 20 percent in weak hours, never below a floor of around 20 percent, hold while the ad set is in learning, and restore to baseline afterwards. It is a budget manoeuvre you approved, not a calendar.
Does Ralph write ad schedules to Google Ads?
Not today. On Google, Ralph shows the hour-of-week evidence and recommends an ad schedule with bid adjustments. Applying it is done by you in Google Ads. We would rather say recommend-only than imply a write that does not exist.
Can weather change my ad schedule in Ralph?
No. Weather is a demand signal in the morning brief. It does not write the hour grid and it does not spend. Weather and the clock run on different timescales, which is explained in weather-based dayparting vs weather-triggered ads.
What does Ralph do if there is not enough data?
Nothing, on purpose. If the hourly history is too thin to show a confident pattern, there is no proposal. Three weeks of ads and a dozen conversions is not a heatmap, and a tool that invents one is guessing with your budget.
Which plan includes dayparting?
The Free plan can see the hourly view and the proposal. Growth and above can approve and apply: a native schedule on lifetime Meta budgets, or the throttle on daily budgets. Details are on the pricing page.
If the hours are the problem: Ralph reads sixty days of your hourly results, proposes the switch-off with floors, and applies it on Meta after you approve. On Google he recommends the schedule. See Ralph · How Ralph dayparts · Docs