The local niche math, inverted
The objection arrives before the idea is finished: a city of 300,000 will never produce a million views. True, and the wrong number to look at. What decides growth is that audience divided by the number of channels chasing it. In a global niche like personal finance or true crime you are one of several thousand; in news for your own city you are one of two or three, and the other two usually quit by month four.
That ratio shows up exactly where the algorithm looks. A thumbnail about a road the viewer drives every morning is not competing with a generic headline, it is competing with nothing. Retention behaves the same way, because someone watching a report about their own neighbourhood finishes it for reasons unrelated to production polish. And the subscriber who followed the city where their children study does not leave when the format changes, since what they subscribed to was the place.
The ceiling is worth stating honestly rather than dressing up. If one percent of a 300,000 person city watches one video a month, that is 3,000 views, arithmetic on an assumption rather than a promise. A local channel runs on frequency and returning viewers, not on one video that travels, which makes it far more predictable than the global niche. It is also the exact shape automation handles well: many small pieces, published on time, every day.
- What decides growth is audience divided by competitors, not audience alone.
- A global niche puts you against thousands, a city puts you against two or three.
- Click through rate rises when the thumbnail is a street the viewer knows.
- One percent of a 300,000 city watching once a month is 3,000 views, as arithmetic.
Where the money actually comes from, and it is not AdSense
AdSense is the smallest line in a local news channel, and pretending otherwise is how people talk themselves out of the niche. News sits in a lower RPM band than finance or software and swings hard by country: estimates commonly land between $1 and $4 per thousand views in the higher paying markets and well under a dollar elsewhere. Against 3,000 views that is a few dollars a video. If advertising were the whole model, the objection would be correct.
What makes the local niche work is direct sponsorship from businesses in the same city. The bakery, the gym, the dealership, the real estate agency and the law firm already buy radio spots and newspaper inches, and they want reach inside a five kilometre radius that no global channel can sell them. The format is a monthly package rather than a per view rate: a mention at the top, a card on screen, a link in the description and one Short a week. Prices are negotiated locally and vary enormously by city, so any figure you read is an estimate.
The structural point is that a monthly package does not depend on RPM at all. A handful of sponsors at a modest monthly rate can outperform the AdSense of a channel ten times larger, and that money arrives whether or not a video travels. What sponsorship demands is that the channel exists every single day. By hand that is eight portals, writing, recording, editing, a thumbnail and an upload before lunch, seven days a week; from an approved calendar with the feeds already wired in, the bulletin happens anyway.
- AdSense on news is a low RPM band and varies hugely by country.
- Local sponsorship is sold as a monthly package and ignores RPM entirely.
- Sponsors buy frequency, so the channel has to exist every day.
- By hand that is four to six hours before lunch, daily, unpaid at the start.
The RSS feeds every city already publishes
This niche is workable at all because a city produces structured, dated, public information every day without anyone asking. Regional news portals are the obvious layer and most still expose a feed. Underneath sits the layer nobody reads: the city hall official gazette, the council agenda and minutes, civil defence and weather bulletins, public security reports, public works and traffic notices, utility outage announcements, the school calendar, the events agenda and the local football club.
Those two layers behave in opposite ways, and that is what makes the mix work. The portals are fast and duplicate each other constantly, because six of them republish the same agency copy under different headlines. The official sources are slow and dry and almost nobody opens them, which is exactly why a bulletin explaining what the council approved on Tuesday has no competition at all. Six to ten sources is a normal working set for one city, weighted so the official ones anchor the facts and the portals supply the pace.
This is mechanical work that has to happen at a fixed hour, which is what automation is for. FalconVid reads the feeds you connect, applies the passing filters that keep an automated channel out of trouble, duplicate and syndication, shelf life, substance and opinion, and turns whatever survives into the calendar. You approve that calendar once, and from then on the feed supplies the topics and the videos come out on schedule. And if the blank page is the hard part, the modeller shows which channels in a niche already monetise.
- Portals are fast, duplicate each other and republish the same agency copy.
- Official gazette, council minutes, civil defence and public works have no competition.
- Six to ten sources is a normal working set for one city.
- The calendar is approved once and the feed supplies the topics after that.
What actually stops people: sensitive facts, duplicates and the line between reporting and opinion
Local news carries categories global content does not. Crime, health, elections and municipal politics have the most local pull and the most ways to go wrong, and the failure is almost never the fact, it is the sentence asserting something nobody confirmed. Naming a suspect who has not been charged, stating a cause of death, presenting a rumour about a candidate as established: those produce takedowns and complaints in a place where everyone knows everyone. The discipline is attribution inside the sentence, so the video says according to the police bulletin rather than stating it flatly. Defamation and privacy rules are national law and this is general orientation, not legal advice, but attribution is good practice everywhere.
The second problem is mechanical. Six portals publishing the same agency story is one topic, not six videos, and a channel that does not deduplicate produces a week of near identical bulletins the audience notices immediately. The third is shelf life: a local story is worth something for hours and nothing by the weekend, the opposite of the evergreen niches. A road closure covered two days late is not a small mistake, it is a video that should not exist.
The fourth is the line between reporting and commenting, and it is the one that costs money. Opinion on crime, health or politics is what makes local advertisers nervous and what platform policy scrutinises hardest, while a bulletin that reports and attributes is boring in exactly the way that keeps a sponsor for two years. FalconVid writes from the source you connected with the attribution attached, and the Studio lets you watch the first version and shorten the intro, swap media or cut a line before anything publishes. Reviewing a four minute bulletin is a completely different amount of work from producing one. The other half of that caution is what you take from the source itself, and the rules on using news articles in your videos covers where the real claims start.
- Attribution goes in the sentence: according to the bulletin, the council published.
- Six portals on the same agency story is one topic, not six videos.
- A local story is worth something for hours and nothing by the weekend.
- Reporting keeps sponsors, opinion is what makes them and the platform nervous.
The publishing window: same day, or the video should not exist
Everything above collapses into one operational requirement. Local news has to go out on the day it happens, so the cycle is daily and production has to fit inside a morning. By hand a single bulletin is four to six hours from opening the first portal to hitting publish: reading, selecting, writing, recording, editing, building a thumbnail and uploading. That is why most local channels that start well are dead by month three, and it is not a discipline problem, it is arithmetic.
FalconVid runs that entire chain with AI specialists working in parallel rather than in sequence, a researcher, a scriptwriter, a narrator, an editor and sound design at once, and a video is ready in up to 30 minutes. Because they run in parallel, more than one video can be in production at a time: 2 simultaneous generations on Starter, 5 on Pro, 10 on Business, 25 on Agency and 50 on Scale. In practice that is the morning bulletin and the afternoon update building together instead of queueing.
What you approve is the calendar, not each script. You set the channel up once, connect the feeds, approve the schedule, and from then on the videos are researched, written, narrated, edited, captioned and published to YouTube, Instagram, TikTok, Rumble and Facebook without you being in the loop. Karaoke captions are on in every plan and carry no credit line of their own, which matters here because most of that audience watches on a phone with the sound off.
Every creation feature is on every plan: what changes is volume, the number of channels, simultaneous generations, the dedicated server from Pro up, the AI Senior Analyst from Pro up with a 7 day trial on Starter, and support. That is what decides whether a local channel survives its own third month, because the version that dies is always the one that depended on somebody having a free morning.
- By hand a daily bulletin is four to six hours, which kills most local channels by month three.
- AI specialists run in parallel and a video is ready in up to 30 minutes.
- Simultaneous generations: 2 on Starter, 5 on Pro, 10 Business, 25 Agency, 50 on Scale.
- You approve the calendar once, not each script, and karaoke captions are on in every plan.
The credit math: why local news is the cheap niche to run
A 12 minute video in economy mode costs 1,008 credits. Starter is $47 for 15,000 credits a month, which is 14 videos in economy mode, roughly 3 hours of finished video. Nobody runs a whole month in a single mode, so the honest way to read Starter is around 10 to 12 videos a month mixing economy with one in balanced, where a balanced 12 minute video is 8,676 credits and a premium one is 26,760.
Cost is also not proportional to duration, because the fixed part does not repeat, and this is where a bulletin channel wins. The pipeline base is 16 credits, research and script is 307 and an economy thumbnail is 479, so 802 credits are paid once no matter how long the video runs, which is 80 percent of a 12 minute economy video. A 12 minute video is 1,008 credits, a 30 minute one around 1,320, and three separate 10 minute videos around 2,920. For a local channel that means a six minute bulletin costs almost the same as a twelve minute one, so you plan by number of videos and publish fewer, fuller bulletins.
The last piece is that this format genuinely does not need the premium engine, which is what makes the niche cheap to operate. A local bulletin is a map, a photograph, a chart and a voice that reports clearly, so economy at 1,008 credits is correct and premium at 26,760 would be money spent where no viewer is looking. In FalconVid a 9:16 Short render is 5 credits, an economy image is 63 and a premium one is 214, so a channel publishing daily plus Shorts stays inside a Starter or Pro budget. This niche rewards volume and speed, precisely the two things that cost the least.
- 12 minute video in economy mode: 1,008 credits. Balanced 8,676, premium 26,760.
- Starter at $47 with 15,000 credits is 14 economy videos, realistically around 10 to 12 mixing modes.
- Fixed costs do not repeat: 30 minutes is around 1,320 credits, three 10 minute videos around 2,920.
- A Short render is 5 credits, so a daily bulletin plus Shorts fits a Starter budget.
The natural expansion: the next city, the Short and the second language
Once one city works, the template is the asset. The neighbouring city has the same structure of sources, the same format and the same sponsor categories with completely different feeds, so the second channel is a configuration rather than a new project. That is where the plan stops being about credits and starts being about channels: FalconVid Starter runs 1 channel, Pro at $97 runs 5 on 30,000 credits, which is 29 economy videos and 5 simultaneous generations. Business at $297 runs 10 channels on 95,000 credits, Agency at $597 runs 25, and Scale at $997 runs 50 channels with 50 simultaneous generations.
The same day of material also has a second life vertically. A local bulletin cuts naturally into a 9:16 Short, a road closure, a match result, a weather warning, and the Short render costs 5 credits, reaching the part of a local audience that lives on Instagram, TikTok and Facebook. Each channel keeps its own calendar, identity and language, so the city channels never blur into one another.
Then there is the second language, which matters more in some regions than people expect. Duplicating a finished project into another language pays only the difference, the narration line, roughly 3.6 percent of a 1,008 credit economy video, with 63 languages available. In a city with a large immigrant community that is the cheapest audience expansion any channel has: the same reporting, the same day, in the language the other half of the neighbourhood speaks at home.
So the limit in this article was never the niche, it was the person. One human being can serve one city, badly, until the mornings run out. With production running in parallel, simultaneous generations and 50 channels on Scale, the ceiling stops being how many mornings you have and becomes which plan you are on.
- The neighbouring city is a configuration, not a new project.
- Pro at $97 is 5 channels, 30,000 credits and 29 economy videos, Scale at $997 is 50 channels.
- A 9:16 Short render costs 5 credits and reaches the audience on the phone.
- A second language pays only the narration, around 3.6 percent of a 1,008 credit video.