Sender Score: What It Is and How to Improve Yours

By Kam Low · Updated 2026-08-19

Sender Score

A sender score is a numeric rating of your sending reputation, from 0 to 100, that mailbox providers use as one signal in deciding whether your mail reaches the inbox.

Someone else's number against your own
Vendor score0 to 100one vendor's model of your reputation, read after the fact
Domain health score1.0 = healthycomputed per domain, on your own signals
gated per domain at send time · the part that makes a score operational
nothing acts on this number
A score you can only read is trivia. A score wired into the send decision is management.

What it measures

Sender Score, run by Validity, rates an IP from 0 to 100 based on complaint rates, spam-trap hits, blacklist appearances, and message volume. Higher is better, and above 90 is healthy. It's one reputation signal among several, not the whole story, but a low score reliably correlates with delivery problems. It predicts inbox placement rather than measuring it, so treat it as one reading inside the broader email deliverability diagnosis.

Read it next to your deliverability rate, which is the outcome the score is trying to predict, and against the requirements tracked in email deliverability news, because the thresholds that decide whether a given score is safe keep moving.

George, our CEO, keeps a correction to how these scores are usually read in his Email Marketing Bible: for Gmail, domain reputation matters more than IP reputation, and IP reputation carries roughly a 120-day memory. He recommends a dedicated IP only once a sender passes a million emails a month.

How to improve it

The levers are the usual ones: authenticate with authentication, cut your bounce rate, keep complaints down, and remove disengaged recipients. A low score is a symptom. The fixes are upstream in list hygiene and sending practice, which the email deliverability best practices page sets out in order.

George treats engagement as a virtuous cycle. Decent engagement improves inboxing, which boosts future engagement. The reverse also holds. The more unopened emails you leave sitting in inboxes, the worse your deliverability gets.

He has watched the thresholds from the operator side too. At SmartrMail, the email platform he ran before Nitrosend, senders were flagged once unsubscribe and complaint rates passed 0.5 percent, a level he calls very high against industry averages closer to 0.02 percent.

Suppression is not hypothetical for us either. The final Brevo blast for Obooko, one of George's own projects, came back with 8,730 addresses to suppress: 3,708 hard bounces, 3,904 soft bounces, 1,105 unsubscribes and 13 spam complaints. All of them are excluded before any next send.

One more structural lever: split your sending across subdomains. It isolates reputation between classes of mail, so a big promotional send cannot take receipts and password resets down with it.

Watching a Sender Score dashboard and manually adjusting volume or content when it dips is reactive babysitting. Nitrosend was built for an agent to do that watching instead: tell it in Claude, ChatGPT or Cursor to keep sending, and reputation tracking runs underneath every send with no dashboard for a human to check.

How Nitrosend tracks reputation

Nitrosend grades your spam rate against Gmail's 0.30% block threshold, tracks per-provider placement trends, and gives every domain a health object with a headline state. It's a live view of the reputation a sender score summarises, tied to the specific domain you send on. Keeping transactional email on its own domain is the cheapest way to protect that score, since receipts almost never draw complaints and should not share a reputation with a promotional list. Sends through the email API report into the same per-domain health object.

We report domain health as a reputation score where 1.0 is healthy, and we track failures, bounces, complaints and auto-pauses underneath it, so a number that moves can be explained rather than just observed.

I wrote up how this behaves at volume in Sending Email at Scale: bounce and complaint rates are tracked per domain, and every send is gated on the health of the specific domain it goes out on. If one brand's domain starts bouncing, that domain gets throttled or held while other brands keep sending on their own clean domains.

The deliverability layer flips a domain into an action_needed state the moment its paid spam rate reaches 0.30 percent or it appears on a blocklist. Auto-pausing came from George pushing for it after watching a bouncy account: high bounce, unsubscribe or spam senders get paused automatically instead of burning the shared pool.

The score maps to real accounts. One sender on our managed warmup tenant held a reputation score around 0.99 across 800 plus sends, with bounce rate near 1 percent, zero spam complaints and zero unsubscribes.

It also catches drift early. George watched the bounce rate on his own customer onboarding flow, 596,000 subscribers on it, climb from 2.15 percent to 3.45 percent, past our 3 percent warning threshold. The score was the smoke alarm. The fix was upstream in the list.

FAQ

What is a sender score?

A 0-to-100 rating of your sending reputation based on complaints, spam-trap hits, blacklist status, and volume, used by providers as one delivery signal.

What is a good sender score?

Above 90 is healthy. Below that correlates with delivery problems, usually driven by complaints, bounces, or blacklisting.

How do I improve my sender score?

Authenticate your domain, cut bounces, keep complaint rates low, and remove disengaged recipients. The score follows list hygiene and sending practice.

Related reading

Transactional email, defined

The transactional email guide

Webhooks explained

Email validation