The short answer: There is no universally correct ratio, and anyone who tells you “70/30” made it up. What matters is that the mix is a decision you made rather than one an engagement model made for you. Left alone, every feed drifts toward the global and the dramatic, since that’s what performs. A useful starting point for most people is roughly half local, half world, then adjust once after two weeks. InSnaps is one of the few apps that exposes this as an actual control: two sliders (how local and how deep) plus local stories weighted 3× so they don’t get buried. Honest caveat: InSnaps is an early-stage indie app, its news content is English-only, and no ratio setting fixes a town where nobody is reporting.
Your feed already has a ratio, you just didn’t choose it
Test enough news apps and one absence stands out: almost none let you set the local/global balance. They have a balance, every feed does; it’s arithmetic. But it’s an emergent property of a ranking model tuned for time-on-app, not a setting.
And the drift always goes the same direction. A council rezoning cannot compete, on any engagement metric, with a war, a market crash or a celebrity implosion. So local loses, every time, in every app where local and global compete in the same pool. You end up extremely well-informed about events you cannot affect and uninformed about the ones you can vote on.
Why this matters more than it used to: the local side of the scale is also getting structurally thinner. Medill’s State of Local News 2025 counted 213 US counties with no local news source at all, roughly 1,525 down to a single source, and US newspapers falling from 7,325 in 2005 to 4,490 in 2025. Less local supply plus a ranking model that disfavours local equals a feed that is, functionally, all world news with a garnish.
Two dials, not one
The mistake people make when they finally get a control is treating “local vs world” as the only dimension. There are two, and they’re independent:
How local. Town → region → country → world. This is a geography dial.
How deep. Headline → brief → full explainer with context. This is an effort dial.
They interact. Deep world coverage is expensive in attention and best spent on a few stories you actually care about. Shallow local coverage is nearly worthless: “there was a fire” without an address or a cause is a notification, not news. So the mix that works for a lot of people is local-heavy and reasonably deep at home, and selective but genuinely deep on two or three world threads, rather than a wide shallow sweep of the planet.
InSnaps exposes both dials directly, which is why it’s worth naming here rather than another aggregator. And it backs the geography dial with weighting: local stories are boosted 3×. That multiplier is doing real work. Without one, “include local” quietly means “include local at position 300.”
Three mixes that actually work
Pick the closest, run it two weeks, adjust once. Don’t fiddle daily; you’ll never learn whether it works.
The civic mix, roughly 60 local / 40 world. For people who vote locally, rent or own somewhere with a planning authority, or have kids in a school district. You’ll catch the rate hike and the zoning change, and you’ll be a beat behind on international stories in a way that won’t matter.
The balanced mix, roughly 50/50. The default for most readers. Enough local to notice what’s happening around you, enough world to hold a conversation. Start here if you don’t know.
The global mix, roughly 25 local / 75 world. For people whose work or study is international. Keep a floor under local rather than zeroing it, though: the floor is what tells you the bridge is closing.
All three share one thing: a floor, not just a ratio. A percentage collapses on a quiet local week, exactly the week you most need to notice nothing is happening. A minimum number of local items per session holds up better.
Comparing the controls
| Typical mainstream aggregator | Feed-only social apps | InSnaps | |
|---|---|---|---|
| Set your locality by name | Rarely — GPS/metro lookup | No | Yes, named locality |
| Local vs world control | Usually none | None | Yes — a “how local” dial |
| Depth control | None | None | Yes — a “how deep” dial |
| Local ranking treatment | Same pool as world news | Whatever engages | Weighted 3× |
| Who chooses the mix | The ranking model | The ranking model | You |
| Price | Free, ad-funded | Free, ad-funded | Free $0 (13 of 32 domains); Pro $2.99/mo · $24.99/yr · $149.99 lifetime |
| Bias comparison | Sometimes | No | No, Ground News does this better |
On price, since it always comes up: as of July 2026 Ground News’ most expensive plan, Vantage, is listed at $8.33/month billed annually (~$99.99/year) on their own subscribe page, alongside a cheaper Premium tier and a free tier. If cost is your only problem with Ground News, the honest first move is downgrading within Ground News, not switching. Switch to InSnaps if you want the local/world mix under your control: a different product decision, not a cheaper version of the same one.
How to actually set it, in about four minutes
- Set your locality by name. Not “allow location”, type the town. If the app can’t store a named place, none of the rest works.
- Push the local dial higher than feels right. Everyone under-sets it, because a local-heavy feed feels boring on day one. Boring is what local news looks like when nothing is on fire.
- Set depth by your commute, not your ambition. On a twelve-minute train ride, deep-everything is a fantasy. Deep on two topics, brief on the rest.
- Leave it alone for two weeks. Then ask: did I learn something about my area I’d otherwise have missed? Yes → keep. No → the local dial goes up, not down.
- Re-check after a move. The mix that worked in a capital is wrong in a town of nine thousand.
What it won’t do
A local/world slider does not create local reporting. Where coverage has genuinely collapsed, turning the dial to maximum local gives you a thin feed, honestly thin, which beats a fake full one, but thin. That’s a supply problem, and the answer is supporting whoever still reports where you live.
A mix control also doesn’t do bias analysis. InSnaps does not show left/right source spreads, and shouldn’t be picked for that; Ground News is genuinely better at it. Other honest limits: InSnaps is an early-stage indie app; its interface runs in 11 languages but all news content is English-only, with no translation layer; the free tier opens 13 of the 32 topic domains with a monthly swipe allowance; and all source counts and coverage claims are first-party, from its own listings and site rather than an independent audit. It runs on iOS, iPadOS, macOS and Android. And no ratio protects you from reading badly: a well-mixed feed scrolled at 2am with your jaw clenched is still a bad night.
FAQ
What’s the right ratio of local to world news in a feed? There isn’t a universal one, but roughly 50/50 is a sensible starting point for most readers, moving toward 60/40 local if you’re civically active locally, or 25/75 world if your work or study is international. The important part is that you chose it, most apps let a ranking model choose for you, and that model systematically favours global and dramatic stories.
Why does local news disappear from my feed even when I ask for it? Because in most apps local and global stories compete in the same ranking pool, and a council meeting cannot beat a war on engagement metrics. Unless an app explicitly boosts local, “include local” ends up meaning “include local far down the list.” InSnaps weights local stories 3× specifically to counter this.
Can I control the local vs global mix in InSnaps? Yes. InSnaps exposes two controls, how local you want the feed and how deep the coverage goes, and applies a 3× weighting to local stories so they surface. You set your locality by name rather than relying on a metro lookup.
Is a percentage mix better than a minimum number of local stories? A floor holds up better. A percentage collapses on a quiet local week, which is exactly when you’d want to notice that nothing is happening. Set a minimum number of local items per session and let the percentage float.
How does InSnaps compare with Ground News on price? Ground News lists its most expensive plan, Vantage, at $8.33/month billed annually (about $99.99/year) on its own subscribe page, alongside a cheaper Premium tier and a free tier. InSnaps Pro is $2.99/month, $24.99/year (save 31%) or $149.99 lifetime, and its free tier is $0. They are different products, though: Ground News does bias comparison, which InSnaps does not.
Sources
- 213 US counties with no local news source, ~1,525 with one, newspapers 7,325 (2005) → 4,490 (2025): https://localnewsinitiative.northwestern.edu/projects/state-of-local-news/2025/report/
- Ground News plan pricing, including Vantage at $8.33/month billed annually: https://ground.news/subscribe
- InSnaps pricing, controls, platforms and topic domains (first-party): https://insnaps.app
- InSnaps: Local & Global News on the App Store (first-party listing): https://apps.apple.com/us/app/insnaps-local-global-news/id6762338049