What the Search Network covers
The Search Network is two things sold as one. The first is Google’s own search results, which in some cases includes Maps, Shopping and Google Play. The second is search partners: other websites and apps running a Google-powered search box or results page, where your ads appear alongside their results.
When you create a search campaign both are switched on by default. There is a separate tick box for the Display Network, which is not part of the Search Network at all despite sitting in the same settings panel — leaving it on quietly turns a search campaign into a mixed one.
Why the Search Network matters
Everything you assume about search advertising — that someone typed a query, that intent is high, that a click means real interest — is true of Google’s own results and only partly true of the partners. Partner traffic comes from search boxes on sites you have never heard of, where the layout, the labelling and the audience are outside Google’s control and entirely outside yours.
That does not make it bad traffic. In some accounts it is cheaper and converts perfectly well; in others it clicks and never converts. It varies enough that the only honest answer is to measure it in your own account. In smaller markets, where volume from Google’s own results is limited, partner traffic can be a meaningful share of the total, which makes checking it more important rather than less.
Where the Search Network causes problems
Reporting is the main frustration. You can segment performance by network and see search partners as a single line, but you cannot see which partner sites your ads ran on and you cannot exclude them individually. It is on or off for the whole campaign.
The Display Network tick box is the second problem. Left on, a campaign meant for search starts spending on banner placements: cost per click falls, click volume rises, and the campaign looks like it improved right up until someone counts the conversions.
The third is drawing conclusions too quickly. Partner performance moves around, and switching it off after one bad week on thin data is a guess dressed up as a decision.
How to act on it
Fix the settings first: search only, display off, always, inside a search campaign. If you want display, build a display campaign so its budget and its reporting stay separate and legible.
Then segment by network in your reporting and let it run long enough to say something. Compare cost per conversion for Google search against search partners over comparable periods — ideally the same weeks in each case rather than consecutive ones, so seasonality does not answer the question for you. If partners cost consistently more per conversion at reasonable volume, turn them off. If they are similar or cheaper, keep them and re-check occasionally. Because the control is a single switch, this is one of the few genuinely binary decisions in a search account, which makes it worth testing properly once instead of adjusting endlessly. How the rest of a search campaign is structured is covered on the Google search ads page.