Article
6 Deliverability Metrics Every Sender Should Track (With Formulas)
By Unlimited Verifier Team ·

Summary
Every sender should track six deliverability metrics: hard bounce rate, spam complaint rate, authentication pass rate, domain and IP reputation, inbox placement, and the invalid share of the list. Each metric needs a fixed denominator, an alert level at 1% hard bounces and 0.1% complaints, and a hard stop at 2% and 0.3%.Every sender should track six deliverability metrics: hard bounce rate (we recommend keeping it under 2%), spam complaint rate (Google and Yahoo both cite 0.3% for bulk senders), authentication pass rate, domain and IP reputation, inbox placement, and the invalid share of your list.
Most senders watch opens and "delivery rate" and miss the signals that decide where mail lands. This guide defines each metric, shows how to calculate it without denominator errors, and ends with a weekly routine and a worked example with full arithmetic. For the broader context, start with the email marketing benchmarks pillar.
Quick answer
- Track six metrics weekly: hard bounce rate, spam complaint rate, authentication pass rate, domain and IP reputation, inbox placement, and the invalid share of your list.
- Use two thresholds for each metric: alert at 1% hard bounces and 0.1% complaints, and treat 2% and 0.3% as hard stops.
- Use the right denominator: bounces divided by sent, complaints divided by delivered. Mixing them is the most common reporting error.
- Don't judge deliverability by opens. Apple Mail Privacy Protection inflates them.
- Verify before you send. The invalid share of a new batch is the earliest warning you control before a campaign goes out.
What is deliverability, and why isn't delivery rate enough?
Deliverability is the share of your mail that reaches the inbox. Delivery rate only shows that a receiving server accepted the message, which is a much weaker signal.
A server that accepts mail replies with a 250 code. A refusal is a permanent 5xx failure, and the enhanced status codes in RFC 3463 explain why. A code such as 5.1.1 means the destination mailbox doesn't exist. A 4xx response is a temporary deferral, usually a soft bounce that may succeed on retry.
The gap is where most teams get caught. Imagine an ESP reporting 98% delivered while a meaningful share of that accepted mail lands in spam at Gmail or Outlook. Acceptance tells you the server didn't reject the message. Only placement, complaints and reputation tell you whether anyone saw it.
Bottom line: Delivery rate measures acceptance. Deliverability is placement, so track the metrics that show what recipients and providers actually do.
Which deliverability metrics should every sender track?
Track these six metrics, each with a denominator you write down. The table gives the formula, the alert level, the hard stop, and why each one matters.
| Metric | Formula (denominator) | Alert | Hard stop | Why it matters |
|---|---|---|---|---|
| Hard bounce rate | Hard bounces ÷ sent | 1% | 2% | Reveals dead addresses and how old your list is |
| Soft bounce rate | Soft bounces ÷ sent | Any address soft-bounces twice | Same address fails on 3 sends | Recurring soft bounces usually become hard failures |
| Spam complaint rate | Complaints ÷ delivered | 0.1% | 0.3% | The strongest negative signal at Gmail and Yahoo |
| Authentication pass rate | Messages passing SPF, DKIM and DMARC ÷ sent | Any week-on-week drop | Sending domain fails DMARC | Bulk senders are expected to authenticate, and failures point to misconfiguration or spoofing |
| Domain and IP reputation | Provider-reported category | Any downgrade | Low or Bad | Shows how providers score you, not just one campaign |
| Inbox placement | Seed accounts in inbox ÷ seed accounts tested | Downward trend | Sustained drop | The only direct measure of placement, though it is sample-based |
| Invalid share of list | Invalid addresses ÷ addresses verified | 2% of new signups | 5% of new signups | Leading indicator of next month's bounces |
The invalid-share thresholds are our rule of thumb, not a published standard. Two more metrics help you diagnose problems. Unsubscribe rate (unsubscribes ÷ delivered) is worth investigating above about 0.5% per campaign, again a practitioner rule rather than a platform figure. Verified-age coverage (the share of your active list verified in the last 90 days) shows how much of your list is still an assumption.
Bottom line: Six metrics, each with a fixed denominator and two thresholds, cover the ground that open rates and delivery rates miss.
How do you calculate bounce and complaint rates without fooling yourself?
Calculate bounce rate against everything you sent, and complaint rate against everything that was delivered. Using one denominator for both produces numbers that look fine and hide real problems.
Bounces are failures to deliver, so they belong in the sent total. Complaints only come from mail that reached a mailbox, so they belong in the delivered total. Google's Postmaster Tools frames its spam rate against delivered mail, so use the same base when you compare your figures with it. For the basic formulas, see how to calculate email bounce rate.
Two more pitfalls catch experienced teams:
- Small sends are noisy. One complaint on a 500-recipient send is a 0.2% complaint rate. Judge complaint trends over rolling four-week windows or on sends above 5,000 recipients.
- Blended averages hide segments. A healthy overall number can conceal one imported list that is failing badly. Break bounces and complaints down by acquisition source before you act.
The worked example below shows a blended complaint rate that looks acceptable while one segment runs at 1.47%.
Bottom line: Bounces divide by sent, complaints divide by delivered, and every rate should be broken down by source before you act on it.
Why are open rates unreliable for deliverability?
Open rates are unreliable because Apple Mail Privacy Protection, introduced with iOS 15 in 2021, preloads tracking pixels whether or not a person reads the email. Opens from Apple devices therefore don't reflect human attention.
That makes opens a poor deliverability signal in two ways. They can overstate engagement for lists heavy on Apple devices, and they tell you nothing about spam placement. Clicks, complaints and placement tests are stronger indicators. To compare the two engagement metrics, read click-to-open rate vs click-through rate.
Bottom line: Judge deliverability with clicks, complaints and placement tests. Treat opens as a rough engagement trend at most.
Where can you measure domain and IP reputation?
You need at least two sources, because each one shows a different slice of your reputation. The table compares the main options by what they show and where they fall short.
| Source | What it shows | Coverage | Main limitation |
|---|---|---|---|
| Google Postmaster Tools | Domain reputation (High, Medium, Low, Bad), spam rate, authentication results, delivery errors | Gmail only | You must verify your sending domain in DNS, and low-volume senders may see no data |
| Microsoft SNDS | IP-level status and complaint data for Outlook.com and Hotmail traffic | Microsoft consumer mail | Works per sending IP, so a shared IP pool blurs your own signal |
| ESP dashboard | Per-campaign sent, bounces, complaints, unsubscribes | Your campaigns only | "Delivered" and "bounce" are defined differently by each platform, so check the definitions |
| Seed and inbox placement tests | Where seeded test accounts land, by provider | Sampled providers | A sample, not your real audience, and pricing varies by vendor |
| Pre-send list verification | Valid, Risky and Invalid counts, plus catch-all, disposable and role flags | Your list before sending | Measures address quality only, not opens, clicks or placement |
For pre-send verification, Unlimited Verifier sorts each address into Valid, Risky or Invalid. It checks syntax, domain, MX records and the SMTP mailbox, detects catch-all domains, and flags disposable and role-based addresses, so you can see why an address landed in Risky. Results export as CSV, Excel or JSON, which makes a monthly log of counts easy to keep. Its email verification API handles single checks and batches of up to 10,000 emails per request, and API access is a Pro-plan feature.
When you compare best email verification tools, judge bucket design and accuracy claims, not price alone. A cheap score helps less than clear Valid, Risky and Invalid buckets you can act on.
Bottom line: Use Postmaster Tools for Gmail reputation, SNDS for Microsoft, your ESP for campaign results, and a verifier for list quality. No single source covers all four.
How do you track list health before you send?
Track the invalid share of every new batch, the catch-all share, the disposable and role-address share, and verified-age coverage. Together they show next month's bounce risk while you can still act on it.
Verification reduces hard bounces, but it is not a consent check. A verifier cannot tell whether someone asked for your mail, and it cannot reliably identify spam traps, so a clean-looking list can still generate complaints. That is why verification is hygiene, not a complaint fix. The email verification compliance and hygiene approach treats the two as separate jobs.
Put verification at the points where addresses enter your system:
- At signup, use real-time email validation to keep the invalid share of new contacts low. For the API pattern, see the email verification API and automation overview.
- On every import, verify before the file reaches your ESP. This matters most for purchased or event leads. B2B teams should also read email verification for sales prospecting.
- On a schedule, follow email list maintenance best practices. Fast growth outpaces manual checks, so compare your intake against the average email list growth rate.
To spot-check a single address, such as a high-value contact that came back Risky, use the free single-email checker.
Bottom line: Invalid share is your leading indicator. If it rises week over week, a signup form or an import source is breaking.
How do you set up a deliverability tracking routine?
Build a routine around a fixed weekly cycle, one shared sheet, and written thresholds. Each step below prevents a specific failure.
- Write your definitions down. Record which denominator each rate uses and which ESP field maps to each count. Without this, two people will report different bounce rates from the same data.
- Pull campaign data weekly. Export sent, hard bounces, soft bounces, complaints and unsubscribes for every send, so you catch trends before they become a review.
- Log provider data weekly. Record Postmaster domain reputation, spam rate, authentication results and SNDS status. Reputation moves slowly, so a weekly log shows the direction of travel.
- Verify every import and signup batch. Log Valid, Risky and Invalid counts with the source name, so a failing segment is identifiable later.
- Review rolling four-week averages. Check segment-level rates first, then the blended total. This stops one odd send from triggering a reaction.
- Act on thresholds, not hunches. At 1% bounces or 0.1% complaints, investigate that segment. At 2% or 0.3%, pause it until you've fixed the cause.
Bottom line: A written definition, a weekly export and fixed thresholds turn deliverability from a vague worry into a routine.
How do you build a deliverability tracker in a spreadsheet?
You can run this with any spreadsheet and the exports your ESP and verifier already produce. It takes about an hour to set up.
- Create columns. Add Date, Campaign, Source, Sent, Hard bounces, Soft bounces, Complaints, Unsubscribes, Delivered, Hard bounce %, Complaint % and Invalid %.
- Enter the campaign data. Copy the counts from your ESP report for each send. Set Delivered as Sent minus Hard bounces, which keeps the complaint denominator correct.
- Add the rate formulas. In Hard bounce %, divide Hard bounces by Sent. In Complaint %, divide Complaints by Delivered. Format both as percentages with two decimal places.
- Tag every row with a source. Use labels such as "Organic signup," "Trade show import" or "Legacy CRM." Source tags are what expose failing segments.
- Add verification counts. Create a free account if you don't have one, then run each import through Upload and verify a list. Enter the total, Valid, Risky and Invalid counts, and calculate Invalid % as Invalid divided by total.
- Set conditional formatting. Turn a cell amber at 1% hard bounces or 0.1% complaints, and red at 2% or 0.3%.
- Add a rolling average. Calculate the four-week average of each rate per source, using AVERAGEIFS or a pivot table.
- Archive everything. Save each export. Unlimited Verifier keeps results for 24 hours on the Free plan and 30 days on Pro, so store exports yourself if you need a longer history.
Bottom line: Eight columns and a handful of formulas produce a tracker that catches segment-level problems the ESP dashboard hides.
What does a worked deliverability example look like?
The figures below are hypothetical, modelled on a mid-size retailer that sends one campaign to a 120,000-contact list. The arithmetic is exact. For sector norms, see ecommerce email marketing benchmarks.
The send: 100,000 contacts came from organic signups, and 20,000 came from a legacy import. The campaign produced:
- Hard bounces: 300 organic and 1,260 imported, for 1,560 total.
- Complaints: 80 organic and 275 imported, for 355 total.
Blended rates:
- Hard bounce rate: 1,560 ÷ 120,000 = 1.30%. That sits between the alert and hard-stop lines, so it looks manageable.
- Delivered: 120,000 − 1,560 = 118,440.
- Complaint rate: 355 ÷ 118,440 = 0.30%, right on the 0.3% hard stop.
Segment rates: The blended numbers hide the real picture.
- Organic hard bounce rate: 300 ÷ 100,000 = 0.30%.
- Imported hard bounce rate: 1,260 ÷ 20,000 = 6.30%.
- Organic complaint rate: 80 ÷ 99,700 delivered = 0.08%.
- Imported complaint rate: 275 ÷ 18,740 delivered = 1.47%, about 18 times the organic rate.
Verification and cost: The 20,000 imported contacts fit on one Pro list, since Pro allows up to 25,000 emails per list. Standard verification is unlimited on every plan. This run uses 20,000 of Pro's 25,000 monthly deep verifications, leaving 5,000 for spot-checks. At $29 a month, that is $1.45 per 1,000 deep verifications actually used (29 ÷ 20) or $1.16 per 1,000 across the full allowance. On the $25 annual rate, the same figures are $1.25 and $1.00. For how to compare plans, see email verification pricing.
Suppose the run returns 14% Invalid (2,800), 9% Risky (1,800) and 77% Valid (15,400). The totals reconcile to 20,000.
Projection: Suppress the 2,800 Invalid and 1,800 Risky addresses, and mail the imported segment only to its 15,400 Valid contacts. Assume the organic segment repeats its 0.30% rate and the cleaned import bounces at a hypothetical 1%:
- Organic bounces: 100,000 × 0.30% = 300.
- Import bounces: 15,400 × 1% = 154.
- Blended: (300 + 154) ÷ (100,000 + 15,400) = 454 ÷ 115,400 = 0.39%.
The insight: Verification cut the projected bounce rate in the imported segment from 6.30% to about 1%. It did nothing for the 1.47% complaint rate. A verifier can't see consent, so that rate has no reason to fall unless you stop mailing people who never opted in. Fix it with re-permissioning or by suppressing the segment, and use the organic 0.08% as your benchmark for healthy. For the consent side of this, see email verification compliance and hygiene and the FTC's CAN-SPAM compliance guide.
Bottom line: A blended number can look acceptable while one segment is far outside your thresholds. Break every rate down by source, and remember that verification fixes bounces, not consent.
Which thresholds should you use when sources disagree?
Sources differ, so pick the threshold that matches the job: the platform's review trigger tells you when a provider will act, and your own alert tells you when to fix something first.
| Source | Figure cited | Notes |
|---|---|---|
| Amazon SES documentation | A 5% bounce rate can trigger review, and 10% can pause sending | Its reputation material also references a 0.1% complaint figure, so check the current page before quoting it |
| Google sender guidelines | Keep spam rate below 0.10%, and never let it reach 0.30% | Applies to bulk senders, defined as those sending 5,000 or more messages a day to Gmail |
| Yahoo sender requirements | Keep complaint rate below 0.3% | Sits alongside authentication requirements |
| Our recommendation | Alert at 1% hard bounces and 0.1% complaints; hard stop at 2% and 0.3% | No major provider publishes a 2% bounce figure, so that ceiling is ours, set well below SES's 5% review trigger |
The sources disagree in two places. For bounces, SES's 5% is a platform review trigger, while 2% is our ceiling because waiting for a provider to act is too late. For complaints, Google's aim of 0.10% is stricter than Yahoo's 0.3% limit, so we use 0.1% as the alert and 0.3% as the hard stop. For more on how complaint levels compare across sectors, see what is a good spam complaint rate and email marketing benchmarks by industry 2026.
Bottom line: Use 1% and 0.1% to trigger investigation, and 2% and 0.3% to stop a segment. Don't plan around a platform's upper limit.
What are the most common deliverability tracking mistakes?
Most mistakes come from definitions and denominators, not from missing data.
- Dividing bounces by delivered. This understates the bounce rate, because delivered is smaller than sent. Use sent.
- Dividing complaints by sent. This understates complaints, because many sent messages never reached a mailbox. Use delivered.
- Judging a campaign on one send. Small sends swing widely, so one complaint can move the rate by a lot. Use rolling windows.
- Reading blended averages only. A healthy total can hide a 1.47% segment, which is exactly what the worked example shows.
- Treating opens as proof of placement. Apple's pre-fetching inflates opens, so they can't confirm the mail reached the inbox.
- Assuming verification fixes complaints. A valid address can still belong to someone who never opted in. Complaints need consent work.
- Losing your history. Results expire after 24 hours on Free and 30 days on Pro, so archive exports on your own schedule.
- Checking Postmaster once. Domain reputation changes week to week, so log it on a fixed cadence.
- Ignoring repeated soft bounces. A soft bounce that recurs across sends often becomes a hard failure, so treat the third one as a hard stop.
Bottom line: Fix the denominators first, then add segment breakdowns, then add verification. Most tracking problems are solved by the first step.
How we compared (and limitations)
This guide is based on public documentation: RFC 3463, Google and Yahoo sender guidelines, Google Postmaster Tools, Microsoft SNDS, Apple's published description of Mail Privacy Protection, and Amazon SES's published thresholds. We haven't run controlled tests against any provider's filters, and platform thresholds change without notice.
The Unlimited Verifier details come from the company's published plans. Its stated accuracy for deep verification is 99.5%, which is the company's own figure, not an independent benchmark. The worked example is hypothetical, and its arithmetic reflects the plan limits described above. Opens, clicks and inbox placement are outside what a pre-send verifier measures, so pair verification with the reputation and campaign sources in this guide.
Bottom line: Use these thresholds as a starting point, and check them against your own data each month.
Worked example: weekly deliverability check
| Step | Calculation | Result | Status |
|---|---|---|---|
| Hard bounce rate | 800 ÷ 50,000 sent × 100 | 1.6% | Above 1% alert, below 2% hard stop |
| Delivered | 50,000 − 800 hard − 300 soft | 48,900 | Base for complaints |
| Spam complaint rate | 120 ÷ 48,900 delivered × 100 | 0.25% | Above 0.1% alert, below 0.3% hard stop |
| Action | Suppress hard bounces, verify list, review complaint source | Fix before the next send |
Frequently asked questions
What is a good hard bounce rate?
Keep hard bounces under 1% of sent mail, which is the alert level in this guide. Treat 2% as a hard stop, where you should pause sending to that list and verify addresses before continuing. Divide hard bounces by total sent, not delivered, so the number reflects list quality.
What spam complaint rate is too high for Gmail and Yahoo?
Google and Yahoo both set 0.3% as the threshold for bulk senders, and this guide treats 0.1% as the alert level. Keep complaints below 0.1% where possible. Measure complaints as a share of delivered messages, not sent, so the rate reflects how recipients actually responded.
How do you calculate spam complaint rate correctly?
Divide the number of spam complaints by the number of messages delivered, then multiply by 100. For example, 120 complaints on 48,900 delivered messages equals 0.25%. Using sent as the denominator understates the rate, which is one of the most common reporting errors.
What is the difference between hard and soft bounces?
A hard bounce is a permanent failure, such as a nonexistent mailbox, usually returned as a 5xx code with an enhanced status like 5.1.1 under RFC 3463. A soft bounce is temporary, often a 4xx deferral, such as a full mailbox. Any address that soft-bounces repeatedly should be treated as a likely hard failure.
What is authentication pass rate and which records matter?
Authentication pass rate is the share of sent messages that pass SPF, DKIM, and DMARC checks. Track it weekly, since any drop usually signals a DNS change, a new sending service, or spoofing. If the sending domain fails DMARC, treat it as a hard stop until the cause is fixed.
How do you measure inbox placement?
Inbox placement is usually estimated with seed accounts, which are test mailboxes you control across major providers. Divide the number of seed accounts that receive the message in the inbox by the total number of seed accounts tested. Seed results are a sample, so compare trends over several sends rather than one test.
Why aren't open rates reliable for deliverability?
Apple Mail Privacy Protection preloads images, which inflates open counts and makes opens a weak signal of whether mail reached the inbox. Use complaints, bounces, authentication, and placement tests instead. Opens can still show engagement trends, but they should not be the main deliverability metric.
What is the invalid share of a list and why check it before sending?
The invalid share is the percentage of addresses on a list that fail verification, such as nonexistent or disposable mailboxes. Checking a new batch before sending gives you an early warning you control, since invalid addresses become hard bounces once you mail them. Remove or suppress flagged addresses before the campaign goes out.
How often should you review deliverability metrics?
Review the six core metrics weekly, and check any campaign's bounces and complaints within 24 to 48 hours of sending. Investigate immediately if any metric crosses its alert level. Monthly reviews are too slow to catch a reputation problem before it compounds across several sends.
Does a high delivery rate mean good deliverability?
No. Delivery rate only shows that a receiving server accepted the message, not that it reached the inbox. Mail can be accepted and still land in spam, which is why placement, complaints, and reputation give a more accurate picture of deliverability.
Cut Bounces and Protect Your Sender Reputation With Unlimited Email Verification
Every invalid address you send to hurts your sender reputation and your inbox placement. Verifying your list before each send means fewer bounces, healthier deliverability, and a predictable cost. With Unlimited Verifier, standard verification is unlimited on every plan, with no per-email fees.
Unlimited Verifier checks syntax, domain, MX records and the SMTP mailbox. It also detects catch-all domains and flags disposable and role-based addresses. Export a CSV from your email platform or CRM, upload and verify the list, then re-import the clean results. You can also automate checks with the email verification API.
| Free | Pro | |
|---|---|---|
| Price | $0 | $29/month ($25/month billed annually) |
| Lists | 25 a month, up to 150 emails each | Unlimited, up to 25,000 emails each |
| Deep verifications | 750 a month | 25,000 a month |
| API access | No | Yes (100,000 emails a month) |
| Result retention | 24 hours | 30 days |
Why teams choose Unlimited Verifier
- Unlimited standard verification on all plans, including Free, with no per-email fees.
- Catch-all detection so you know which domains accept every address.
- Clear Valid / Risky / Invalid results, exportable as CSV, Excel or JSON.
- An open-source verification engine you can inspect on GitHub.
Unlimited Verifier states 99.5% accuracy for deep verification.
Create a free account and verify your next list before you hit send.
Want to test a single address first? Try the free email checker.