Google Ads search partners: switch off by default or leave on for reach?

I'd turn search partners off by default, especially on smaller accounts. Google blends them into the same campaign, so the overall CPA can look fine while a chunk of spend goes to sites that barely convert. The usual defence is reach, but on a budget of a few hundred a month that reach is mostly low-intent clicks you then have to clean up with negatives.

What I'd do is start with partners off and let the search terms report show what's working. Then test partners in a separate campaign with its own budget, so the comparison is clean. Has anyone seen partner traffic hold up at lower spend levels?

I'm not sure the separate campaign test holds up at a few hundred a month. Smart bidding needs roughly 30 conversions a month to learn well, and splitting a small budget in two usually leaves both campaigns under that, so you end up comparing two campaigns that are both still learning. You can get a clean comparison without splitting anything. Segment the existing campaign by Network and you'll see Search Partners against Google-owned results on the same keywords, budget and bids.

Agreed, the segment view is the cleaner way to read it. Just remember the Network setting applies at campaign level, so the segment is only a read, not a control. Partner clicks come out of the same bids, which means if partners look expensive you can't just lower them. Turning the setting off for the whole campaign is your only real lever.

Since that's a blunt switch, compare cost per conversion over at least four weeks, not day to day. With a few hundred a month, a partner segment might only have five or six conversions in that window, so look at cost per click and the search terms behind those clicks alongside the conversion numbers before you decide.

I wouldn't lean on cost per conversion here. At five or six conversions, one extra or missing lead swings CPA by roughly 15 to 20 percent, so four weeks gives you a direction, not a verdict. I'd use the partner search terms and click quality as the tiebreaker instead.

Agreed that four weeks at five or six conversions is a direction, not a verdict. Whether partner search terms are a fair tiebreaker comes down to how many conversions the account gets per month, since that decides whether conversion data can settle this at all.