How the Display Network works
Publishers sell space on their websites and apps through Google, and advertisers buy it through the same Google Ads account they use for search. When a page loads, an auction runs in the background for the slot, and the winning ad appears beside whatever the person came to read. No search took place, so relevance has to be inferred rather than read from a query.
Google infers it in several ways, and you choose which. You can target audiences — people with a stated interest, people in a life stage, or people who have already visited your site. You can target contexts, meaning pages about a topic or pages containing certain keywords. Or you can name specific placements: individual websites, apps, YouTube channels or Gmail. Most accounts use a mix, and the more of these you stack, the narrower the delivery becomes.
Why the Display Network matters
Its clicks and impressions are far cheaper than search, because nobody on the network has expressed a need. That trade is worth making for two jobs. The first is remarketing: showing something useful to people who already visited, where the intent came from an earlier visit rather than the placement. The second is awareness in markets where few people search the category yet, which is common for new services and for smaller markets such as Nepal.
It also gives reach that search cannot. Search volume has a ceiling set by how many people type the query; display volume is limited mainly by budget and frequency. When a search campaign is already at high impression share, display is one of the few places left to grow.
Where the Display Network goes wrong
Waste is the usual complaint, and it is nearly always a placement problem. Left unchecked, ads drift onto low-quality sites, auto-refreshing pages and mobile games where clicks are accidental rather than intended. The clicks are real, they are cheap, and they convert at close to nothing.
The second failure is measurement. Because display rarely earns the last click, people either declare it useless or, worse, count every view-through as a win. Both readings are wrong, and both lead to bad budget decisions. The third is creative: a single banner shown to the same person day after day stops registering long before the reporting shows it.
How to act on it
Start with remarketing rather than cold targeting, because it is the version most likely to pay back while you learn. Exclude mobile app inventory and game categories at the start, then review the placements report regularly and exclude sites that spend without converting. Cap frequency so one person is not shown the same ad endlessly.
Keep the display budget and reporting separate from search so neither flatters the other, and judge display on assisted conversions and cost per lead over a longer window. If the aim is simply to bring back people who nearly enquired, a tightly built display advertising setup with strong exclusions will usually beat a broad interest campaign.
