How much does SendGrid cost in 2026?

Two products, two price lists, one brand

This is the thing to understand before looking at any number, and it is the single most common source of budget surprises with SendGrid.

SendGrid sells two distinct products that happen to share a name:

This also explains why SendGrid appears in wildly different price comparisons. Against Postmark or Amazon SES it is the Email API being compared. Against Mailchimp or Brevo it is Marketing Campaigns. Comparing the wrong one produces a number that is off by a factor of several.

A note on the brand. SendGrid was acquired by Twilio and is now presented as Twilio SendGrid. The pricing pages have moved to twilio.com, and sendgrid.com redirects there. Nothing about the product changed, but bookmarks and older links break.

Email API: the transactional product

Priced on how many emails you send per month, with no contact database in the pricing model. This is what you buy if your email is generated by your application.

Volume tiers run through the selector at 3000, 50 000, 100 000, 300 000, 700 000, 1.5M, 2.5M and 5M+ emails per month. SendGrid does not publish a flat table of what each tier costs; the prices appear as you select a volume.

Marketing Campaigns: the newsletter product

Priced on two axes at once, contacts stored and emails sent, which makes it harder to model than a single-axis competitor.

Contact tiers: 100, 5000, 10 000, 20 000, 50 000, 100 000, 200 000, 500 000+.

Send tiers: 6000, 15 000, 30 000, 50 000, 60 000, 100 000, 150 000, 250 000, 300 000, 500 000, 1M, 2.5M+ per month.

The free tier is the most generous piece of SendGrid. 2000 contacts and 6000 monthly emails beats every free plan we have documented: MailerLite gives 250 subscribers and 2500 emails, AWeber gives 500 and 3000, Mailchimp gives 250 and 500. If you want a free tier to actually run a small list on, this is the largest one available.

The two free offers are not the same thing

SendGrid publishes free access on both products and they behave differently, which trips people up.

The Email API free allowance of 100 emails a day works out to roughly 3000 a month, but it is a daily cap rather than a monthly one. That matters for transactional traffic, which is bursty by nature. A signup spike or a batch of order confirmations can exhaust a day while your monthly total stays tiny.

Why a daily cap is harder to live with than a monthly one

The Email API free allowance is 100 emails per day, not 3000 per month. Those are not equivalent, and the difference matters specifically for transactional traffic.

Marketing email is scheduled, so its volume is smooth and predictable. Transactional email is triggered by user behaviour, so it arrives in bursts: a product launch, a press mention, a batch of order confirmations after a promotion. A monthly allowance absorbs bursts. A daily allowance does not.

What that means in practice. Do not size the free tier on your monthly volume. Size it on your worst day. If your signup flow or your order volume can spike tenfold on a launch, the free plan will fail on exactly the day it matters most, and transactional failures are visible to customers in a way a delayed newsletter is not.

Marketing Campaigns takes the opposite approach, metering 6000 emails per month with no daily ceiling published. For that product a burst is fine. The two free tiers therefore fail in different ways, which is a reasonable design given what each product carries.

Overage rates, published in detail

SendGrid publishes per-email overage rates by plan and tier, which is unusual and useful. It also reveals an anomaly worth knowing about.

The rate does fall properly at higher volume: $0.0008 at Pro 700K and $0.0006 at Pro 1.5M, which is less than half the Essentials 50K rate. Overage economics improve with scale, just not monotonically across the Essentials to Pro boundary.

The gap between the two products is the real story

Email API overage runs $0.0006 to $0.0013 per email. Marketing Campaigns overage runs $0.0023 to $0.0075 per extra email or contact.

At the extremes that is a twelvefold difference for sending an email through the marketing product rather than the API. Even comparing like for like at the cheap end, marketing overage starts at nearly double the most expensive Email API rate.

The practical implication. If you run both products, the plan you overshoot matters enormously. Overshooting the API is cheap; overshooting Marketing Campaigns is not. And because Marketing Campaigns meters contacts as well as sends, you can trigger the dearer rate through list growth alone without sending a single extra email.

Add-ons: validation, testing credits, extra IPs

The testing credits scale sensibly. Thirty credits at $18 is $0.60 each; a hundred at $40 is $0.40. On the Marketing Campaigns side, two thousand at $800 is also $0.40, so the discount flattens rather than continuing to improve. Buy the hundred-credit bundle and stop; there is no volume advantage beyond it.

Should you run both here, or split providers?

Once you accept that SendGrid is two products, an obvious question follows: if you are paying twice anyway, why not buy each half from whoever does it best?

The argument for splitting is mostly financial, and it is strong at scale. At 500 000 transactional emails a month, SES costs about $50 in send fees. Nothing in the SendGrid Email API range approaches that.

The argument against is operational, and it is stronger than it looks. Running two providers means two authentication setups, two deliverability surfaces to monitor, and two places to look when mail stops arriving. It also means that if your marketing sends damage your domain reputation, you have to have deliberately separated the sending domains for the isolation to actually work. Most teams that split do not, and then get the cost of two vendors without the benefit.

Which product do you actually need?

Choosing wrong here costs more than choosing the wrong plan, because the two products have different pricing axes and very different overage economics.

What five real accounts pay

Because SendGrid is two products, a realistic bill often has two lines. These scenarios use the published starting rates and overage figures.

Read the fourth row carefully. It is the most common real configuration and the one people budget wrong. Two products, two starting rates, and both volumes well above the entry tier, so the actual bill is meaningfully higher than the $34.95 the two starting prices suggest. Price each product separately in its own selector.

Rows one and two are the flip side and worth stating plainly: SendGrid gives away more than most competitors at the bottom. A side project sending a couple of hundred transactional emails a month and a newsletter under 2000 contacts both cost nothing, indefinitely.

How SendGrid compares on price

Two tables, because comparing the wrong product against the wrong competitor is the mistake this page exists to prevent.

Email API against transactional providers

On raw cost per email, Amazon SES is roughly six to thirteen times cheaper than SendGrid overage and nothing here competes with it. What SES does not give you is deliverability tooling, an interface, support, or validation. SendGrid Email API sits where it always has: more expensive than raw infrastructure, considerably cheaper than a marketing platform, with a published dedicated IP at a price you can actually see.

Marketing Campaigns against marketing platforms

Basic at $15 is the most expensive entry point in this group, but it is the only one whose free tier is large enough to run a real list on, and the only one that does not force you to buy send volume proportional to your contact count. For a large list mailed infrequently, the two-axis model genuinely saves money. For a small list mailed often, it does not.

Is the included email validation worth choosing Pro for?

Pro bundles 2500 validations a month and Premier 5000. Validation checks whether an address is real before you send to it, which protects your bounce rate and therefore your sender reputation.

The question is whether 2500 is a useful quantity. It depends entirely on where you validate:

So treat it as a signup-flow feature, not a list-hygiene tool. If your use case is continuous validation of new addresses at a modest rate, the Pro allowance genuinely covers it and the feature is worth counting toward the price. If you are cleaning an inherited list, you will exhaust it immediately and the rate beyond is not published.

What you cannot price in advance

Who should pay for which plan

Frequently asked questions