Netlify Doubled the Credit Cost of Bandwidth and Made Seats Free. If You Run a Traffic-Heavy Site Solo, You Got the Bad Half of That Trade.
Netlify's April 14 changelog did three things at once, and the framing you have probably seen picks one of them. Bandwidth went from 10 to 20 credits per GB. Compute went from 5 to 10 credits per GB-hour. Web requests went the other way, from 3 credits per 10,000 down to 2. At the same time the Pro plan dropped per-seat charging entirely, so $20 a month now buys 3,000 credits for a team of any size, where before it bought one seat.
Four months of billing cycles have run since. This is a good moment to do the arithmetic properly, because the loudest takes call it a straight price hike and that is not accurate. It is a redistribution, and it happens to redistribute away from exactly one profile: the person running a traffic-heavy site alone.
The trade, stated plainly
Netlify moved money from metered resources to seats. If your costs were dominated by people, you won, and you won big. Previously Pro was $20 per seat, and that included active Git contributors, which is the detail that used to make open-source-adjacent teams furious. Now you can invite owners, developers, publishers, contributors and reviewers without a seat charge.
If your costs were dominated by traffic, you lost, and the loss is a clean 2x on the two meters that matter most.
Solo operators are the pure case of the losing profile. I have never had a seat problem. There is one of me. Every dollar of my hosting bill has always been bandwidth and compute, which are precisely the two lines that doubled. The generous read is that I was subsidised by teams paying seat fees for years and the subsidy ended. That is probably true and it does not make the invoice smaller.
Worth keeping in view: web requests got 33% cheaper. If your workload is many small requests rather than large asset transfers, your outcome is better than the headline. That is a real part of the change and it gets left out of every summary I have read.
The arithmetic that matters
Netlify's own changelog gives you $20 for 3,000 credits on Pro and 20 credits per GB of bandwidth. Divide and you get 150 GB of bandwidth per month included on Pro, if bandwidth is the only thing you consume. Before the change the same 3,000 credits would have covered 300 GB.
The free tier is commonly reported at 300 credits, which I have seen in several pricing roundups but could not confirm in Netlify's own changelog, so treat it as secondary. If it is right, 300 credits at 20 credits per GB is 15 GB of bandwidth in a month before your project pauses.
Fifteen gigabytes sounds like a lot until you put images in it. A blog with a 2 MB page weight hits 15 GB at roughly 7,500 pageviews. That is not a viral month. That is a decent Tuesday if something of yours lands on Hacker News.
The number I would actually go and get is your own. Pull the last three months of bandwidth from your Netlify dashboard, multiply by 20, add compute at 10 per GB-hour, and compare against your credit allowance. That takes five minutes and replaces every estimate in this post, including mine.
The legacy plan detail nobody mentions
If you are on a Legacy Free, Legacy Starter or Legacy Pro plan, none of this touched you. Netlify's impact table says so explicitly, three separate rows, no changes with this update.
That is genuinely important and it is missing from most of the coverage. A meaningful number of long-time Netlify users are grandfathered and have been reading angry threads about a pricing change that does not apply to their account. Before you spend a weekend migrating hosts, check which plan you are actually on.
It also creates a trap in the other direction. If you are on a legacy plan and something tempts you to switch to a credit plan for a feature, you are giving up the grandfathering permanently. Netlify's docs treat plan changes as one-way in that respect. That is worth knowing before you click.
This is not only a Netlify story
The pattern I actually care about is broader. Across 2026, several hosts that built their reputations on generous free and flat tiers moved toward metering. Netlify went credit-based. Cloudflare is reported to have introduced a bandwidth cap on free custom domains around June. Vercel moved to credit-metered build machines. I have not verified the Cloudflare and Vercel details to the same standard as the Netlify changelog, so weight them accordingly, but the direction is not subtle.
The reason is not villainy. It is that AI-assisted building produced a lot more deployed projects per user, and unmetered generosity does not survive that. A flat free tier is a bet that most accounts stay small. When every account can spin up ten projects in an afternoon, the bet stops paying.
What that means practically is that the era of picking a host once and never thinking about the bill is over for anyone with real traffic. Not because any single host got greedy, but because the whole category repriced.
What I would actually do
Do the five-minute calculation above before anything else. Most solo projects are nowhere near these limits and the correct action is to close the tab and get back to work. Anxiety about hosting costs is one of the most reliable ways I know to spend a day producing nothing.
If you are over, the highest-leverage fix is not switching hosts. It is putting a CDN and sane cache headers in front of your assets and compressing your images, because bandwidth is the meter that doubled and images are almost always the reason bandwidth is large. Halving your page weight halves the line item, and it works on any host, which means it survives the next repricing too.
If you are still over after that, then compare hosts, and compare them on the meters you actually consume rather than on the sticker price. A plan that is cheaper per month and more expensive per GB is not cheaper for you.
And if you are on a legacy plan: do nothing. Enjoy it. Set a reminder to re-read this if Netlify ever sunsets it.
Where I could be wrong
The strongest counter-argument is that calling this a loss for solo operators ignores that the free and Personal tiers exist precisely for us, and that a person doing 15 GB a month is not really the customer being priced here. Fair. If your project is small, the rate change is theoretical. My complaint is specifically about the middle: enough traffic to exceed a free tier, not enough revenue to make $20 versus $60 irrelevant. That band is real but it is narrower than my framing implies.
I should also flag that I am reading a pricing change through my own bill, which is a biased instrument. Someone running an agency with eight contributors on Netlify got a large, unambiguous discount from the same changelog I am grumbling about. Both things are true, and "vendor repriced in a way that helped teams and hurt individuals" is a more honest headline than "vendor raised prices."
Author
Lukas
@lukcombinator