What you are actually buying
Connected TV means an ad inside streaming content on an internet-connected television. Ad-supported tiers of the major services, free ad-supported channels, and the app layer on smart TVs all sell inventory the same way: by the impression, targeted by audience and geography.
This is a different transaction from traditional television. You are not buying a market and a daypart from a station rep. You are buying households — specific ones, in specific counties, watching specific kinds of content — through the same kind of auction that sells display inventory.
The practical consequence is that the minimum viable buy collapsed. A campaign that used to require tens of thousands of dollars to be worth trafficking now runs sensibly at a few hundred dollars a week.
The actual numbers
Connected TV is priced on CPM — cost per thousand impressions. In most local markets you should expect roughly $20 to $40, with the low end on broader, less premium inventory and the high end on live sports and premium originals.
At $25 CPM, $1,000 buys 40,000 impressions. In a county of 200,000 people, that is meaningful frequency against a targeted slice of households rather than a token presence.
Compare that honestly: paid social runs $8 to $20 CPM, so streaming is more expensive per impression. It is also fifteen to thirty seconds of full-screen, non-skippable attention with sound on, which no feed placement delivers. Cheaper impressions are not automatically better ones.
Why it works for local businesses
Two reasons, and neither is the one people expect.
The first is credibility. Appearing on television — even streaming television — reads as evidence that a business is real, established, and not going anywhere. For home services, medical practices, and anyone asking for a large purchase from a stranger, that perception does measurable work.
The second is that the attention is genuinely uncontested. There is no scrolling past it, no muting it, no competing content in the same frame. Completion rates on connected TV routinely exceed 95 percent, which is a number no other digital channel can claim.
The creative question
You do not need a production. You need fifteen or thirty seconds that answers four things: who you are, what you do, where you do it, and one reason to act now.
Priorities, in order: clean audio, legible text, a static end card with your name and phone number held long enough to read, and footage that looks like your actual business. Camera quality matters less than any of those. Bad audio will sink a spot that looks fine; mediocre footage with clear narration will not.
Use what you have before you shoot anything. Job-site footage, a walkthrough of your shop, a technician explaining what a repair costs — all of it performs, and all of it costs nothing.
How to measure it honestly
There is nothing to click, so click-based measurement tells you nothing. Measure these instead, over the flight and the two weeks after it:
Search volume on your own business name. Direct traffic to your site. Total inbound calls. And blended cost per booked job across your whole program — not per channel, because attribution will hand streaming’s credit to whatever the customer touched last.
The most common mistake is running four weeks, seeing no in-platform conversions, and concluding it failed. Six weeks minimum, and read the blended number.
When not to buy it
Skip it if your total budget is under about $1,500 a month — you need enough left over to capture the demand streaming creates, and search should get funded first.
Skip it if your offer is unproven. Streaming is efficient at telling a lot of people something; make sure the something works before you amplify it.
And skip it if nobody answers your phone. This is not a joke — credibility-driven channels generate calls, and a business that misses half of them is buying awareness for a competitor.