Sender Score Improvement Guide: Understanding Validity's Reputation Metric and How to Move It

Sender Score, maintained by Validity (formerly ReturnPath), is a 0-100 reputation score assigned to sending IP addresses based on a rolling 30-day window of email sending behaviour data. It is one of the oldest email reputation scoring systems still in active use, and while its influence has diminished as Gmail and Microsoft moved toward their own proprietary reputation systems, it remains relevant for two reasons: a subset of enterprise mail servers and spam filters use Sender Score as an input to their filtering decisions, and Sender Score provides a useful leading indicator of reputation health that can catch problems before they manifest as delivery failures.

The most common misconception about Sender Score: that improving it will directly improve Gmail inbox placement. It will not — Gmail uses its own domain reputation system (visible in Postmaster Tools) that is independent of Sender Score. What Sender Score improvement will help with: delivery through enterprise gateways and smaller ISPs that query the Validity reputation database, and as a proxy indicator for the list quality and sending practice improvements that also improve deliverability everywhere else. The actions that move Sender Score up are the same actions that improve email delivery broadly — Sender Score is a diagnostic tool, not the goal itself.

What Sender Score Is and Is Not

Sender Score is a proprietary metric maintained by Validity that aggregates multiple data signals into a single 0-100 score for each sending IP address. Validity collects data from a network of partnering ISPs, mailbox providers, and enterprise mail environments — their network processes data from a claimed 2.5 billion mailboxes. Within that network, they observe complaint rates, unknown user rates (email sent to addresses that don't exist), spam trap hits, and other engagement signals for each sending IP. These signals are combined into the Sender Score using a proprietary formula that Validity does not fully publish.

What Sender Score measures: relative reputation of a sending IP compared to other IPs in Validity's database. A score of 90 means the IP is performing better than 90% of all IPs in the database on the measured signals. It is not an absolute measure — a score of 90 in 2024 may correspond to different absolute complaint rate levels than a score of 90 in 2020, because the database baseline shifts over time.

What Sender Score does not measure: domain reputation (Gmail Postmaster Tools measures this), inbox placement rate (GlockApps and Litmus measure this), DMARC compliance, or domain age. It is one data signal among many, weighted differently by different mail servers that choose to use it. Most large consumer ISPs (Gmail, Yahoo, Microsoft) have their own proprietary reputation systems that they use in addition to or instead of Sender Score. Sender Score's primary relevance is for enterprise environments that use Validity's filtering data and for smaller ISPs that do not operate their own reputation infrastructure.

How Sender Score Is Calculated

Validity does not publish the exact Sender Score formula, but based on observed correlation between Sender Score changes and specific metric changes, the primary inputs are:

Complaint rate (highest weight): The fraction of email sent from the IP that recipients mark as spam within Validity's partner network. Complaint rate is the dominant signal — a sudden increase in complaints from one campaign can drop a Sender Score from 90 to 75 within 24-48 hours. Conversely, sustained low complaint rates over 30 days consistently produce high scores.

Unknown user rate (significant weight): The fraction of email sent to addresses that do not exist — generating 550 5.1.1 "mailbox does not exist" permanent rejections. This is a list quality signal: senders with clean, verified lists have low unknown user rates; senders with old, unverified lists or purchased data have high unknown user rates. A high unknown user rate (above 1-2%) will suppress Sender Score even if complaint rates are low, because it indicates the sender is not maintaining adequate list quality.

Spam trap hits (catastrophic impact): Sending to Validity's spam trap addresses — email addresses seeded in their network that should never receive legitimate email — is the fastest way to drop Sender Score precipitously. A single spam trap hit in the 30-day window can drop a score from 85 to 40. Multiple trap hits in a short period can drop a score to near zero. Trap hits indicate list quality problems severe enough that the sender is contacting addresses that have been expired, sold, or harvested — none of which should happen in a legitimate opt-in programme.

Volume consistency (moderate weight): Erratic sending patterns — large spikes followed by quiet periods — contribute to lower scores. Consistent sending volume over the 30-day window is rewarded. This is the same signal that ISPs use for rate limit calibration — consistent volume is the pattern of legitimate senders, erratic spikes are the pattern of spam operations.

Sender Score Thresholds: What the Numbers Mean in Practice

The threshold that matters most operationally: 80. Mail servers that use Sender Score as a delivery input often apply a threshold around 80 — above which is "acceptable sender" and below which additional filtering applies. The difference between a score of 79 and 81 may be the difference between inbox delivery and spam filtering at those servers. The improvement effort should be focused on getting above 80 as a minimum, and then maintaining 85+ as the operational target.

Complaint Rate: The Primary Sender Score Driver

Reducing complaint rate is the single most impactful action available for Sender Score improvement. Every percentage point reduction in complaint rate within Validity's partner network translates to a measurable score improvement, typically within 7-14 days of the change (the 30-day rolling window means improvements become fully visible within 30 days).

The complaint rate levers, in order of impact:

Remove non-openers aggressively: Subscribers who have not opened email in 90+ days are significantly more likely to mark email as spam when they eventually receive a campaign. Remove them from active sending before they become complaints. The sunset segment approach (detailed in the Klaviyo guide and list hygiene guides on this site) is the systematic implementation of this principle. The counterintuitive truth: a smaller, active list has better Sender Score than a larger list including significant lapsed segments — because the lapsed segments generate complaint rates that the engaged segments do not.

Make unsubscribe genuinely easy: A recipient who cannot find the unsubscribe link, who clicks unsubscribe and reaches a page requiring account login, or who unsubscribes and continues receiving email will mark as spam. The spam mark is a much worse outcome for Sender Score than the unsubscribe would have been. Make unsubscribe links large, obvious, in the footer and in the body where appropriate, functional without login, and processed within 10 calendar days (as required by CEMA and best practice under CAN-SPAM).

Stop sending to complaint addresses immediately: If you are enrolled in FBL (Feedback Loop) programmes at Yahoo JMRP and Microsoft JMRP, you receive complaint notifications for specific addresses. Process these within 24 hours — add to the permanent suppression list immediately. Each complaint address that receives another email after it has already complained generates additional complaint signal in Validity's data.

Unknown User Rate: The Signal Most Senders Ignore

Unknown user rate — hard bounces with the specific code 550 5.1.1 "User does not exist" or similar — is the second most important Sender Score signal after complaint rate. Most senders monitor their overall bounce rate, but few monitor the unknown user rate specifically, and the distinction matters for Sender Score because Validity's system weights 5.1.1 (address doesn't exist) differently from other bounce categories (5.2.2 mailbox full, 5.7.x policy rejections).

# Extracting unknown user (5.1.1) bounces from Postfix logs:
grep "status=bounced" /var/log/mail.log |     grep -E "5\.1\.[12]|User unknown|does not exist|no such user" |     wc -l

# Unknown user rate calculation:
# (5.1.1 bounces in period) / (total emails sent in period) * 100
# Target: below 0.3% unknown user rate
# Alert threshold: above 0.5%
# Critical: above 1.0% — indicates list quality problem requiring immediate list audit

# The sources of high unknown user rates:
# 1. Purchased or scraped email lists
# 2. Very old lists where addresses have aged out (accounts closed, companies defunct)
# 3. Email address typos that were never validated at signup
# 4. Corporate email addresses for employees who left the company
#    (their corporate email closes when they leave)
# 5. Role-based addresses that were decommissioned (team@company.com → deleted)

# Fix: Run the full sending list through email verification
# ZeroBounce, NeverBounce, and BriteVerify all identify:
# - Invalid (will hard bounce as 5.1.1)
# - Valid (will deliver)
# - Risky (deliverable but associated with spam traps or disposable domains)
# - Unknown (cannot determine without sending)
# Suppress "Invalid" immediately; suppress "Risky" if Sender Score is below 80

Spam Trap Hits: The Most Catastrophic Sender Score Event

Validity operates a network of spam trap addresses embedded throughout their partner ISP network. These are email addresses that should never appear on a legitimate opt-in list — they are either former valid addresses that have been expired and repurposed as traps (recycled traps) or addresses that were never valid and only appear in harvested, scraped, or purchased data (pristine traps). Sending to any of these generates an immediate, severe Sender Score penalty that the 30-day rolling window cannot quickly recover from.

The trap scenario most common in practice: a sender acquires a list from a partner, co-registration, or purchased source, adds it to their sending list, and the list contains several Validity trap addresses. The first campaign to the combined list hits the traps; Sender Score drops from 85 to 35 within 48 hours. The sender does not know why — there is no notification from Validity that a trap was hit, unlike the FBL-based complaint notifications from Yahoo JMRP. The only visibility is the Sender Score drop itself, which may not be noticed for days or weeks if Sender Score is not being monitored daily.

Trap hit recovery: once a trap is hit, that address must be suppressed immediately. The damage to Sender Score begins decaying out of the 30-day window automatically, but it takes 30 days for the event to fully age out of the calculation. During those 30 days, avoid sending at high volume to minimise the percentage impact of the trap hit on the aggregate score. Run the full list through email verification to identify and remove any other trap-pattern addresses before resuming full-volume sending. The list acquisition source that produced the trap should be identified and any other contacts from that same acquisition cohort should be treated as suspect — they came from the same data set as the trap address and may have similar data quality issues.

The Sender Score Improvement Protocol

Sender Score vs Gmail Postmaster Tools vs SNDS

Understanding what each tool measures and where it matters prevents the common mistake of using the wrong tool for the wrong problem:

The complete monitoring stack uses all of these, not just one. Sender Score below 80 with Gmail Postmaster Tools showing Good domain reputation indicates the problem is specific to the enterprise environments that use Validity data — likely a list quality issue generating unknown user rates or complaints within Validity's partner network that does not strongly affect Gmail's domain-level signals. Sender Score at 90 with Gmail Postmaster Tools showing Low domain reputation indicates the problem is Gmail-specific — likely a complaint rate issue from Gmail users specifically that does not show in Validity's broader network. The two tools measure different things; both are needed for a complete picture.

Sender Score improvement is ultimately list quality improvement. The score is a lagging indicator of how the sending IP has behaved across Validity's partner network over the past 30 days. The actions that improve the score — removing invalid addresses, suppressing unengaged contacts, processing complaints immediately, eliminating purchased data — are the actions that improve email programme performance holistically. Track Sender Score as a confirmation that the list quality work is having its intended effect, not as the goal itself. A 90+ Sender Score is evidence that your programme is operating well; it is not the objective that makes the programme worth operating.