Scaling a winning Google Ads campaign: step the budget up weekly or jump once?

Say a store has a campaign at 6x ROAS on a handful of products, and the owner wants to triple the budget this week. My instinct is to push back, or at least not do it all at once. Smart Bidding gets unsettled when budget and targets change together, and tripling spend on a campaign with maybe 40 conversions a month drops you into a different auction mix before the model has learned anything. The opposite problem is real too, though. I've seen a store wait three weeks for things to feel stable while its best product sold out twice. So what's the real rule? Is 20 to 30% a week the safe speed, or does conversions per day matter more than the percentage? I'd like to hear how people handle stepped ramps versus one jump, especially on seasonal promos where the window is short.

Conversions per day matters more than the percentage, in my view. A campaign doing 40 conversions a month is only getting 1 to 2 a day, so even a 20% budget step can mean a week of noise before you learn anything. For short promos I'd still step the budget, but I'd also check whether the bid target is changing at the same time, since that's what really shifts the auction mix.

Priya's point on conversions per day holds, but what really decides it is whether the campaign is budget-limited or target-limited. If it's already spending its full daily budget and losing impression share to budget, a jump is usually safe, since the bids aren't changing, only the cap. If it isn't hitting budget, more money won't buy more volume, and stepping up just burns through the learning window.

Budget-limited is the right test, but an unchanged bid doesn't mean an unchanged auction. Once the cap lifts, the campaign starts winning auctions it was losing before, and those tend to be cheaper, weaker-intent queries. So ROAS can slip even though you never touched the target.

What I'd do is raise the budget to roughly what closes the lost impression share from budget, then look at the search terms after two or three days. If the new queries look off, pull the cap back before you touch the target.

Rowan's right about the search terms. One thing to add: look at impression share lost to budget against the actual spend limit before raising the cap. On Search that number is measured across the whole day, so it can hide a campaign that runs dry by mid-afternoon. If the daily cap is being hit at 2pm, bumping it to Rowan's number will mostly buy evening traffic, and for a promo store that's often where the weakest queries live. Watching hourly delivery for a couple of days will show you whether the extra budget lands where you want it.