A consumer packaged goods brand we work with spent three years running the same banner ads through a managed-service network. Decent reach, murky reporting, no idea which placements drove anything. When we moved them to a transparent DSP, we could see within two weeks that 67% of their impressions were landing on made-for-advertising sites their audience wasn't actually visiting. They were paying to be ignored at scale. That's programmatic working against you. The same infrastructure can work the other way.
The point isn't that programmatic is magic. It's that the mechanism is flexible enough to do very different things depending on how you set it up. Understanding what good execution looks like in practice is the only way to know whether your current setup is actually doing its job.
If you want the foundational explanation of how the buying process works, our overview of programmatic advertising covers the auction mechanics and platform types. This post assumes you've got that and want to see what campaigns look like in the real world across specific formats and industries.
Display advertising
Display is where most brands start with programmatic because the format is simple and inventory is everywhere. The banner ads you see on news sites, weather apps, sports scores pages, and recipe blogs are almost all bought programmatically. The advertiser doesn't call the publisher. A DSP bids on each available impression in real time based on whether the user matches the targeting criteria.
A national home improvement retailer uses programmatic display to run three distinct campaign types simultaneously. The first targets people who visited their site but didn't buy, showing product-specific ads for whatever they browsed. The second targets homeowners in specific ZIP codes within 30 miles of a store location. The third runs contextual display on home improvement content across the open web, with no audience data at all.
Each campaign serves a different goal. The retargeting captures intent already expressed on the site. The geo campaign drives local foot traffic. The contextual campaign builds awareness among people in the right mindset who haven't visited yet. They run at different CPMs, different budgets, and get evaluated against different KPIs. That's what a real programmatic display setup looks like.
The thing agencies don't mention about display CPMs. Programmatic display CPMs typically run 25 to 45% lower than direct publisher buys for comparable placements. That cost advantage disappears if you're not monitoring placement quality. Low CPMs on junk inventory are still wasted dollars.
Connected TV (CTV)
CTV is the fastest-growing programmatic format right now, and it's the one where we see the biggest gap between what brands think they're buying and what they're actually getting.
Here's what a CTV campaign actually looks like. A regional healthcare system wants to reach adults 35 to 60 in three metro markets. They work with their DSP to build an audience segment combining age and geography with contextual signals around health and wellness content. That audience gets served a 30-second non-skippable ad while watching streaming content on Hulu, Peacock, and Tubi through a smart TV or streaming device. The ad completes at a 94% rate because there's no skip button. The campaign reports on reach, frequency, and website visits lift from households that were exposed.
That's CTV in practice. It looks like TV but it buys like programmatic display. You're not buying a show. You're buying an audience wherever they happen to be watching. US CTV ad spend reached $33 billion in 2025, up 16% from the prior year, driven largely by advertisers moving budgets away from linear TV where audience measurement has become increasingly unreliable.
The automotive industry adopted CTV aggressively. An auto manufacturer running a new model launch uses programmatic CTV to serve a 30-second video to in-market auto shoppers, identified through third-party purchase-intent data from platforms tracking search and research behavior. The same household might see a different version of the creative based on which vehicle trim they've been researching. That targeting precision is something linear TV can't touch.
Native advertising
Native programmatic ads match the visual format of the publisher content around them. They look like article recommendations, suggested reads, or sponsored posts rather than banner ads. The ad unit says "sponsored" but it doesn't interrupt the content the way a banner does.
A B2B software company selling enterprise HR tools runs native programmatic on business publications and LinkedIn's audience network. The ad looks like an article headline with a thumbnail. It reads "Why HR Directors at Mid-Market Companies Are Rethinking Their Tech Stack." The audience targeting is by job title and company size. When someone clicks, they land on a gated piece of thought leadership, not a product page.
That's the native format working correctly. It earns attention rather than demanding it. The CPMs are higher than standard display but click-through rates are typically two to four times higher. The trade-off is that the creative has to actually be worth reading or the format doesn't work.
Retail media
Retail media is the fastest-growing segment in programmatic right now. The mechanics are the same as standard programmatic, but the inventory lives inside retail environments and the targeting is built on actual purchase data instead of behavioral inference.
A consumer packaged goods brand selling pet food runs a campaign on Amazon Advertising. They target people who've purchased competing pet food brands in the last 90 days, people who've bought their own brand and are likely to repurchase, and a broader category audience of pet owners who've shopped the category but haven't tried their brand. Each audience segment sees different creative and different messaging. Sponsored product placements appear on search results pages. Display ads run across Amazon's owned properties and its off-site network.
The data advantage here is real. Amazon knows who actually bought what. That's not inferred. It's transaction data. Programmatic retail media spend grew 41.7% in 2024 and was on track for another 29.3% gain in 2025, with spending projected to surpass $30 billion that year. The growth is coming from CPG, health and beauty, and household goods brands that have always spent heavily at retail and are now following their audience into the digital shelf.
Walmart Connect and the second-party data advantage
Walmart Connect works the same way. A brand selling sporting goods targets Walmart shoppers who've bought related equipment in the last six months. The placements run both on Walmart's own site and across the broader Walmart DSP network on third-party sites. The CPMs are higher than open-exchange display. The ROI is usually better because the audience quality is higher. You're not inferring intent. You're targeting people who bought.
Digital out-of-home (DOOH)
DOOH is the format that surprises most advertisers when they realize it's in the programmatic ecosystem. Digital billboards, gas station screens, transit display boards, airport panels, and retail signage are all available for automated buying through DSPs.
A quick-service restaurant chain uses programmatic DOOH to push lunch promotion ads during the midday window on weekdays, only on screens within two miles of their locations. The campaign triggers only during those windows. Outside of them, the impressions don't buy. No manual insertion orders. No negotiating screen by screen. The DSP handles it in real time.
The more interesting use cases involve condition-based triggers. A weather app data provider sells a trigger that activates DOOH ads when local temperature drops below 40 degrees. A hot soup brand uses this to push ads on transit screens in cold weather only. The impression logic is automated.
Audio programmatic
Spotify, Pandora, iHeart, and podcast networks all sell audio inventory programmatically. An advertiser targeting runners uses exercise and fitness audience segments to serve a 30-second audio ad during workout playlists on Spotify. The targeting is behavioral and contextual at the same time.
Podcast programmatic is more limited but growing. Dynamic ad insertion lets advertisers place audio spots into podcast episodes in real time, served to different listeners based on geography or demographics even within the same episode. A financial services company targeting high-income households can serve their ad only to podcast listeners who match that profile, even on a general-interest show.
Programmatic by industry
The formats are consistent across industries. What changes is how the targeting logic gets built and what success looks like.
| Industry | Primary Format | Targeting Approach | What Success Looks Like |
|---|---|---|---|
| Retail / E-commerce | Display + Retail Media | Site retargeting, purchase intent, proximity | ROAS, cost per purchase, store visit lift |
| Automotive | CTV + Display | In-market shopper data, model-specific intent | Dealer visits, test drive leads, VDP views |
| B2B / SaaS | Native + Display | Job title, company size, account lists (ABM) | MQL volume, pipeline influence, CPL |
| Healthcare | CTV + Native | Condition-based segments, HCP targeting | Patient appointments, script lift, site visits |
| Financial Services | Display + Audio | Life event data, income proxies, in-market signals | Application starts, account openings, CPL |
| QSR / Food & Bev | DOOH + Mobile | Proximity, daypart, weather triggers | Foot traffic lift, visit frequency |
Healthcare deserves a specific note. Reaching healthcare professionals programmatically requires specialized data and inventory providers. Doceree, MedData Group, and a handful of other platforms hold NPI-linked segments that let pharma and medical device brands reach physicians, nurses, and pharmacists by specialty across medical trade sites and general web properties. The targeting precision is much higher than consumer programmatic because the universe is small and the data is verified. Healthcare programmatic spending doubled between 2021 and 2024 as regulatory clarity improved.
What programmatic can't do
It's worth being direct about the limits because the category gets oversold.
- Programmatic can't fix a bad offer. Serving a mediocre product ad at high frequency to the right audience still produces mediocre results. The automation finds the audience. The creative and the landing page still have to close.
- Programmatic can't guarantee brand-safe placements on open exchange. Without an active inclusion list or private marketplace deals, your ad might appear next to content you'd never want to associate with your brand.
- Programmatic can't replace a full-funnel strategy. Display retargeting captures in-market demand. It doesn't create it. If your top-of-funnel is weak, retargeting a small pool of site visitors hits a ceiling fast.
- Programmatic doesn't self-optimize well enough to run on autopilot. We've seen accounts where nobody reviewed placement reports for months. Made-for-advertising sites, low-viewability inventory, and bot traffic don't disappear on their own. Someone has to manage the blocklists.
The bottom line on programmatic examples
The most effective programmatic campaigns we manage share one thing. They're built around a specific audience action, not just reach. The retailer targeting shoppers within a mile of a store wants a store visit. The B2B company running ABM wants a demo request from a specific account list. The measurement framework matches the objective before the campaign launches.
When we see programmatic campaigns underperform, it's almost always because the targeting was too broad, the creative wasn't differentiated by audience segment, or nobody was actively managing placement quality. The platform doesn't prevent any of those problems. You have to.
If you want to understand the platform infrastructure that powers these campaigns, read our breakdown of how a DSP works. For a look at what a well-run programmatic program costs and includes, our programmatic service page covers that directly.
Programmatic advertising accounts for over 90% of all US digital display spending as of 2025. That number gets cited as if it's a sign of the format's effectiveness. It's really just a sign of how thoroughly it replaced direct buying as the default purchasing mechanism. Effectiveness comes from how you use it, not from the fact that you're using it at all.