The pitch usually arrives with a nice deck. Your commercial, running on Hulu, Roku and the rest of the streaming apps, shown only to homeowners in the zip codes you serve. A monthly report showing how many people saw it and how many of them later visited your website. And, almost as an afterthought, an offer to draw a digital fence around your biggest competitor's showroom and follow everyone who walks in with your ads for the next month.
It sounds like television at a digital price, with digital targeting and digital proof. Some of it is exactly that. Some of it is something else wearing the same name, and the proposal will not tell you which is which.
This is not an argument against streaming TV or geofencing. Both are real, both can work for the right business, and both have become far more affordable than most owners realize. It is an argument for knowing what you are buying before you sign, because this is one of the corners of advertising where the product on the proposal and the product that actually runs can be two different things.
Three Products Get Sold Under One Name
Most pitches blend three different things together. Separating them is the first step to judging any of them.
Streaming TV. You will hear it called CTV (connected TV) or OTT (over-the-top, meaning delivered over the internet rather than through a cable box). Either way, it means your video ad plays inside a streaming app on a television: the ad break in a show on Hulu, Tubi, Peacock or a local station's streaming channel. It is usually full screen and often cannot be skipped. You pay per thousand impressions, the same way you pay for most online display ads, and the price sits well above ordinary online video because the screen is bigger and the viewer is sitting down.
Programmatic buying. This is the machinery, not a place. A demand-side platform, or DSP, is software that bids in real time for ad space across thousands of apps and websites at once. Streaming TV is one of the formats it can buy. So are banner ads on news sites, pre-roll video before online clips, and ads inside mobile games. When someone says "programmatic," ask what formats are actually in the plan, because the word covers everything from a premium TV spot to a banner at the bottom of a recipe blog.
Geofencing. This is location targeting. The seller draws a boundary around a real place, a trade show, a competitor's lot, a neighborhood. When a phone with location services turned on enters that boundary, its advertising ID is added to an audience. You can then show ads to that phone, across apps and websites, for weeks afterward. Practitioners commonly describe a follow-up window of about thirty days.
These three do different jobs. Streaming TV makes you visible on the biggest screen in the house. Programmatic is a way of buying, and its value depends entirely on what gets bought. Geofencing is a way of choosing who sees the ad based on where they have been. A proposal that wraps all three into one monthly fee, with one blended result, makes it impossible to know which part is earning its money.
Streaming TV Is Where the Audience Went
Before the warnings, the honest case, because it is a strong one.
People did not stop watching television. They moved how they watch it. Nielsen's measurement of US TV viewing put streaming at 49.0% of all television time in July 2026, nearly half of everything watched on a TV. The shows a local business used to reach through a cable spot at 6:30 p.m. are now watched on demand, on apps, whenever people feel like it.
And the price of getting in has collapsed. Buying television used to mean a sales rep, a production budget and a minimum that ruled out most small businesses. Now Disney's self-serve tool, which sells ads on Hulu, starts at $500 per campaign. Vibe, a self-serve streaming TV platform, advertises campaigns from $50 a day. Google Ads lets a video campaign target TV screens as a device, which means an ad you already run on YouTube can show up on the living-room television with no separate minimum at all.
So streaming TV is no longer something only large brands can buy. That is the good news. It is also why so many people are now selling it to you, and why it pays to understand what they are selling.
Ask Where Every Impression Actually Ran
Here is the first thing to know, and it comes from the sellers themselves. In a webinar for home service businesses, an agency that sells streaming TV ads described regularly reviewing new clients' reports from other vendors and finding that what had been billed as streaming TV was mixed with pre-roll and display. Pre-roll is the short video before online clips. Display means banner ads. Both are cheaper to buy than a streaming TV spot, and both can be packaged into a "streaming TV" line at streaming TV prices. Their words: not all of it is created equal.
The difference matters beyond price. A thirty-second ad playing full screen during a show someone chose to watch is a different experience from a six-second video before a clip on a phone, or a banner nobody noticed. If the proposal says TV, the report should prove TV.
The second thing is fraud. Streaming TV attracts it for the same reason it attracts advertisers: the prices are high. It is the same problem we covered in paying for clicks from people who are not people, at a higher price per impression. DoubleVerify, one of the large ad verification companies, reported in May 2026 that it detected 140% more streaming TV fraud schemes in the first quarter of 2026 than a year earlier, and that campaigns running without fraud protection saw a fraud rate of nearly 9%, against less than 1% in protected campaigns. One scheme it documented ran ads through screensaver apps on streaming devices, producing impressions that nobody needed to be watching. Its fraud lab also warned that the belief that buying directly from a publisher is fraud-free "is not the case."
None of this means streaming TV is a scam. It means you should ask for the same evidence you would expect from any other ad you pay for:
- A placement report by app or channel. Not "premium streaming inventory," but the actual names of the apps and channels where your ad ran.
- A device breakdown. What share of impressions played on a television, and what share on phones and computers.
- A completion rate. How many of the ads were watched to the end.
- The name of the verification company. Who is checking that real people on real screens saw the ads, and whether that check is included in your price.
A good seller will have all of this ready. A seller who cannot produce it is telling you something important.
A Walk-In Report Is Not Proof the Ad Worked
The report will almost certainly show results. The question is what those results actually measure.
Streaming TV and geofencing sellers commonly prove their value in three ways. The first is matching: the ad played on a TV in a household, and later a device on the same home internet connection visited your website, so the visit gets credited to the ad. The second is branded searches: after the campaign started, more people searched for your business by name. The third, for geofencing, is walk-ins: phones that saw your ad were later detected at your location.
All three are real signals, and all three share the same weakness. They count people who saw the ad and then did something, without asking whether they would have done it anyway. The household that visited your site might have been a returning customer. The branded searches might have risen because it was your busy season. The phone that walked into your store might belong to someone who comes in every week. We have written about how ad platforms tend to overstate their own results, and about why you will never know exactly which ad made the sale. Streaming TV and geofencing reports have the same problem, often with less transparency about how the matching works.
The fix is a test that measures the difference the ads actually made, the same holdout approach that post recommends, applied here by geography. Split your service area in two. Run the ads in one group of zip codes and hold them back in a comparable group. Run it long enough for a genuine difference to show up, then compare leads, calls and sales between the two groups, using your own records rather than the vendor's report. If the zip codes with ads produce clearly more business than the zip codes without, the ads are doing something. If they look the same, the matched visits were mostly people who were coming anyway.
Some self-serve platforms now build holdout testing in. If you are buying from an agency or a local station, ask for it before you sign, not after the first report arrives.
Geofence the Places You Would Be Proud to Stand
Geofencing is where the pitch gets most exciting, and where it most needs a clear head.
It has genuinely good uses. If you exhibit at a home show, a trade show or a community event, putting a boundary around it and following up with the people who attended is a smart, focused use of a small budget. A home builder can reach people walking a new development. A business can reach people at its own location, or in the parking lot of a large retailer its customers also visit. The appeal is obvious: you are reaching people because of where they have physically been. It is retargeting by place instead of by website visit, and the same discipline about how long to follow someone applies.
It is not expensive per impression. One practitioner, in a widely watched 2021 explainer, put typical costs at roughly $2 to $10 per thousand impressions, with about a $1,000 monthly budget as a practical floor. But the audience inside a boundary is often small, so the real question is whether enough of the right people pass through it.
The legal and ethical lines matter more than most sellers mention. New York and Washington have both banned geofencing around health care facilities for advertising purposes, and the Federal Trade Commission has taken repeated action against companies that collected and sold location data showing visits to sensitive places. In its case against the data broker Gravy Analytics, announced in December 2024 and finalized in January 2025, the FTC banned the company from selling or using location data tied to medical facilities, places of worship, schools, shelters and similar locations. If you run a health, wellness or medical business, geofencing needs particular care, and the targeting ideas in a generic pitch may not be legal where you operate. Even where it is legal, some places should simply be off the table. One geofencing practitioner draws the line at hospitals, funerals and churches, and so should you.
That points to one question every geofencing buyer should ask: where does the location data come from, and did the people in it agree to share it? Location data reaches ad platforms through apps and data suppliers. Some of that data was collected with clear consent. Some, as the FTC's cases show, was not. You do not want your business name attached to the second kind.
Who Streaming TV Actually Fits
One of the more candid things in this research came from a sales trainer teaching local media companies how to sell more streaming TV. His advice to the sellers was that the product works wonders for the right advertisers, but "certainly not everyone can benefit," and that they should stop wasting time on the ones who cannot. It is worth knowing which group you are in before the rep does.
Streaming TV tends to fit a business when a few things are true at once:
- The purchase is considered, not impulsive. Roofing, remodeling, HVAC replacement, dental implants, legal services, a car, a pool. People think about these for weeks, which gives a TV ad time to matter.
- You serve a defined area. Targeting by zip code is what makes a small streaming budget viable. A business that serves everywhere needs a much bigger budget to be seen often enough anywhere.
- Search is already working. Streaming TV makes people aware of you. It works best when there is already something in place to catch the demand it creates. If your Google Ads and website are not converting yet, that is where the money should go first. We have laid out which ad channel a small business should start with, and streaming TV is rarely it.
- You have, or can afford, a real commercial. The same sales trainer called the lack of a commercial one of the biggest obstacles to closing these deals. A weak ad on the biggest screen in the house does more harm than good.
Then do the arithmetic, because this is where small budgets get stretched thin without anyone saying so. Take a price of $30 per thousand impressions, which sits in the middle of what streaming TV sellers commonly quote. A $1,500 monthly budget buys 50,000 impressions. A TV ad usually needs to be seen several times before it registers, so at five views per household, that $1,500 reaches about 10,000 homes a month. For a business serving a few zip codes, that might be meaningful coverage. For a business serving a metro area, it is a rounding error, and no amount of clever targeting changes the math.
This is also why the minimum spend matters as much as the price per impression. The cost of reaching a market is set by the market, not by the platform, a point we have made about Google Ads minimum spend that applies here just as much.
Six Questions Before You Sign Anything
When the proposal arrives, you do not need to become an expert in ad technology. You need answers to six questions, in writing:
- Where exactly will my ads run? Which apps and channels, and what share on television screens versus phones and computers. Ask to see a sample report from a current client.
- Who verifies that real people saw them? The name of the verification company, and whether it is included in the price.
- How will we know it worked? Ask for a holdout test by zip code, measured against your own leads and sales, not only matched visits or walk-ins.
- What am I committing to? The minimum spend, the contract length, and how you get out if the numbers do not move.
- Who makes the commercial, and what does it cost? Get the production cost separately, and ask to see examples they have made for similar businesses.
- Where does the audience or location data come from? Especially for geofencing, and especially if you are in health care.
A seller who welcomes these questions is probably selling you the real thing. A seller who gets vague on the first two is probably selling you something cheaper under a better name. The same test applies to anyone managing your advertising: we have written about how to tell if the people running your ads are any good, and the answer comes down to the same thing. Good people show you what they did and what it produced. They do not ask you to take the reach number on faith.
The Screen Changed. The Rules Did Not.
Streaming TV and geofencing are genuinely new ways to reach people, and for some businesses they will be worth every dollar. But the rules for judging them are the same rules you should apply to every ad you buy. Know what you are paying for. See where it ran. Measure the difference it made against your own numbers, not the seller's. And do not let the size of the screen, or the cleverness of the targeting, stand in for proof.
If the answers to the six questions are good, try it, with a holdout, on a budget you can afford to learn from. If they are not, you have lost nothing but a meeting.
