Netlify Just Doubled the Cost of a Credit on April 14 — Read Your New Bill Before Renewal

Quietly, on April 14, Netlify changed what a credit buys.

Bandwidth went from 10 to 20 credits per GB. Compute went from 5 to 10 credits per GB-hour. A new "web requests" line item showed up at 2 credits per 10K. Same 300 credits/month allocation on the Pro plan. Same dollar price.

Half the bandwidth. Half the compute.

Form submissions did go free, which is nice if your product is Mailchimp. Form submissions did not go free if your product is anything else.

If you migrated to Netlify in the last year because Vercel's bill scared you, your renewal is about to be a different conversation. And the broader pattern — credits-based pricing where the credit silently devalues — is the 2026 deployment-platform UX equivalent of shrinkflation.

The exact change

Before April 14: bandwidth at 10 credits/GB, compute at 5 credits/GB-hour, no separate web requests line item, form submissions metered at 1 credit each.

After April 14: bandwidth at 20 credits/GB (2× cost), compute at 10 credits/GB-hour (2× cost), web requests newly metered at 2 credits per 10K, form submissions free.

Same plan structure. Same $19/month Pro for 300 credits, same $99/month Business for 1,500 credits. The headline "no price change" is technically true and substantively misleading. Your effective per-GB price doubled.

Who actually pays for this

The indie dev hosting a small SaaS that does ~50GB/month of bandwidth and ~30GB-hours of compute.

Pre-April 14 math: 50GB × 10 credits + 30GB-hours × 5 credits = 650 credits. Pro plan covers 300, you pay overage on 350 credits. At ~$0.11 per overage credit, that's roughly $40/month of overage on Pro.

Post-April 14 math: 50GB × 20 credits + 30GB-hours × 10 credits = 1,300 credits. Pro plan still covers 300, you pay overage on 1,000 credits. At the same overage rate, that's roughly $80/month of overage on Pro.

For a $200 MRR side project, that's a meaningful gross margin hit. The overage doubled. The plan price didn't change. Read your bill.

The form submissions counter

Form posts dropped from 1 credit each to free.

If you're running a marketing site with a contact form and a newsletter signup form, you net out fine. The bandwidth/compute increase is roughly offset by the form-post savings on a low-traffic content site.

If you're running an actual web app — a SaaS where every page load involves dynamic compute and the form submissions are a tiny fraction of total traffic — you got worse pricing dressed up as a cleanup. The "form submissions free" concession is structurally a marketing-site benefit, not a SaaS benefit.

This is the part that telegraphs intent. Netlify is positioning more deliberately as a marketing-site platform, where the unit economics work, and quietly making itself more expensive for actual web apps, where the unit economics are harder.

The pattern, named

Credits-based pricing is the new deployment-platform shrinkflation.

When a platform converts a clear per-resource bill ("$0.05 per GB of bandwidth") into an opaque credits abstraction ("300 credits per month, where a credit buys variable amounts of stuff"), the per-credit value will move down over time and the headline plan price stays flat.

This is the same pattern as airline points, gym memberships, B2B SaaS "platform fees," and basically any metered service where the abstraction layer between dollars and units is opaque enough to be quietly devalued. The reason platforms reach for credits-based pricing is that it allows price increases without the optics of a price increase. The dollar number stays the same in your invoice, the credit allocation stays the same in your plan summary, only the per-credit value drops — and that drop is a config change, not a pricing announcement.

Plan accordingly. Every credits-based deployment platform is on the same trajectory.

The Cloudflare comparison every solo operator should run this weekend

Same workload, same numbers. 50GB bandwidth + 30GB-hours compute + 100K requests.

On Netlify post-April-14: ~1,300 credits, ~$80/month of overage on Pro.

On Cloudflare Workers Paid: $5/month flat with generous overage. Bandwidth doesn't have a meter on Cloudflare because Cloudflare's infrastructure makes egress effectively free for them. The structural reason is real — Cloudflare's business model is fundamentally different, and that difference shows up as zero bandwidth metering.

The math difference for the modal indie SaaS workload: ~$75/month savings, every month. ~$900/year. The migration cost is real (1–2 weekends for a typical Astro + adapter swap, plus DNS update), but the savings clear that within 60–90 days.

I'm on Cloudflare for this blog and have been for the post-Astro-acquisition period. The bill is consistently under $5/month at low traffic. I'd move to Cloudflare again today, with the acknowledgment below.

The honest counter-take

Netlify's product is genuinely well-loved by people who picked it for the developer experience and Functions surface, not the price.

If you migrated for the DX, this change probably doesn't move you. Netlify's deploy ergonomics, branch previews, edge functions, and split-testing tools are real and the team has built them out steadily. Pricing isn't the only variable.

If you migrated for the price (and a lot of solo devs did, post-Vercel-bill-shock pieces last year), the math changed and you owe yourself a reread. The Cloudflare-Pages-and-Workers comparison this weekend is the right exercise.

My bias: I'm on Cloudflare for this blog and would move to Cloudflare again today, but acknowledge "you're betting on a single vendor's cost structure" is a real critique. Cloudflare's bandwidth-is-free pricing depends on Cloudflare's infrastructure choices remaining favorable. If Cloudflare's economics shift in 2027, the pricing argument shifts with them. The risk profile of vendor concentration is real even when the current pricing is best.

The migration-decision framework

Three questions, in order.

Did you pick Netlify for the DX or the price? If DX, stay. If price, run the numbers.

What's your real workload mix? Marketing site with forms — Netlify post-April-14 is fine. Actual web app — Netlify just got more expensive without giving you anything in return.

What's your migration cost? A typical Astro or Next.js site moves to Cloudflare Pages in 1–2 weekends. A complex Netlify Functions setup with edge functions and middleware is closer to a week. Estimate honestly. Don't guess.

For the modal indie SaaS workload, the math points to Cloudflare. For a marketing-site workload, Netlify post-April-14 is still fine. For a complex Netlify-Functions-heavy setup, the migration cost might exceed the savings — stay and pay the new pricing, or wait until you have time to do the migration properly.

The actionable move this week: read your last three Netlify invoices. Compare the line items. Recompute against the post-April-14 pricing. If the new bill is meaningfully higher, do the Cloudflare comparison this weekend. If it's roughly the same, stay and revisit at renewal.

The credits-based pricing pattern isn't going away. Vercel will probably ship its own credits restructure in the next 6 months, with the same shape — same headline price, less compute behind the abstraction. Pre-empt it by knowing your real-resource costs, not your credit allocation.