Microsoft's Smart Network Data Services (SNDS) is a free, underused tool that gives email senders direct visibility into how Microsoft's mail ecosystem perceives their sending infrastructure. If you send any meaningful volume to Outlook.com, Hotmail, or Live.com addresses and you are not checking SNDS regularly, you are flying blind on roughly 400 million mailboxes worldwide. This guide covers how to set up SNDS, what each metric actually tells you, and how to act on the data before deliverability problems compound.
Most senders discover SNDS after something has already gone wrong. A campaign that used to hit 30% open rates on Microsoft domains suddenly drops to 12%, and someone on the team starts Googling "Outlook spam filter" at 9 p.m. The better approach is to treat SNDS like a vital sign monitor: check it routinely, set internal thresholds, and respond to early signals before they become full-blown reputation crises. Microsoft does not offer the granular feedback loops that some other mailbox providers do, which makes SNDS one of the few windows you get into their filtering decisions.
Getting started requires a Microsoft account and ownership (or authorized control) of the IP addresses you send from. You sign up at sendersupport.olc.protection.outlook.com/snds, request access to your IP ranges, and verify ownership through a process that typically involves confirming you can receive email at the abuse or postmaster address for those IPs. Verification can take anywhere from a few hours to a couple of days. One thing that trips up teams: if you use a shared IP pool from your ESP, you may not be able to enroll those IPs in SNDS yourself. You will need your provider to either grant you access or share the data. This is worth asking about during vendor evaluation, because any ESP that cannot give you visibility into Microsoft reputation data is asking you to trust their deliverability claims on faith.
Once enrolled, the SNDS dashboard surfaces data across several dimensions. The two most important are the traffic summary (showing volume, trap hits, and complaint rates by IP) and the reputation status, which Microsoft categorizes as green, yellow, or red. Green means normal filtering. Yellow means some messages are being filtered more aggressively. Red means a significant portion of your traffic is being junked or blocked. The color coding is simple by design, but the simplicity can be misleading. A green status does not mean every message hit the inbox; it means Microsoft's systems do not currently consider your IP problematic at a macro level. You can still have individual campaign or content-level filtering happening beneath a green IP reputation.
The trap hit metric deserves special attention. Microsoft operates its own network of spam traps, and SNDS reports when your IPs hit them. Trap hits are not always a sign of purchased lists or malicious behavior. They can also indicate list hygiene decay: subscribers who abandoned their Microsoft addresses, which were then recycled into trap addresses after a dormancy period. A single trap hit on a high-volume IP is not necessarily cause for alarm, but a pattern of recurring hits across multiple days signals that your list cleaning process has gaps. The fix is almost always upstream, in how you acquire and maintain addresses, not downstream in trying to suppress specific trap addresses you will never be able to identify.
Complaint rate is the other metric that demands consistent monitoring. SNDS shows you the percentage of messages from each IP that recipients marked as junk. Microsoft's internal threshold is not publicly documented with precision, but operational experience across high-volume senders suggests that complaint rates above 0.3% start triggering increased filtering, and rates above 0.5% put you in serious jeopardy. For context, a well-managed sender to Microsoft domains typically sees complaint rates between 0.05% and 0.15%. If you are consistently above 0.2%, something in your sending program needs to change: frequency, targeting, content relevance, or opt-in process.
| SNDS Metric | What It Tells You | Action Threshold |
|---|---|---|
| IP Reputation (Green/Yellow/Red) | Overall filtering disposition for your IP | Yellow = investigate within 24 hrs; Red = immediate remediation |
| Trap Hits | Messages hitting Microsoft-operated spam traps | Any recurring pattern across 3+ days |
| Complaint Rate | % of recipients marking mail as junk | Above 0.3% = warning; above 0.5% = critical |
| Sample Messages | Example message data from filtered mail | Review for content patterns triggering filters |
SNDS also provides sample message data when your traffic is being filtered. This is gold. Most senders never look at these samples, but they reveal which specific messages triggered filtering, including subject lines, sending domains, and content characteristics. If you notice a pattern (for example, all filtered samples share a particular URL domain or use similar promotional language), you have a concrete starting point for remediation rather than guessing at what went wrong.
One limitation worth understanding: SNDS only reports on traffic to consumer Microsoft mailboxes (Outlook.com, Hotmail, Live.com). If you are sending to corporate Microsoft 365 tenants, that traffic is governed by each organization's own filtering policies, Exchange Online Protection settings, and potentially third-party security gateways. SNDS will not show you data on those deliveries. For B2B senders whose primary audience is on corporate Microsoft 365, SNDS still matters because your IP reputation with Microsoft's consumer network influences how Exchange Online Protection evaluates your traffic to business tenants. The systems are not identical, but they share intelligence.
When you do hit yellow or red status, the remediation playbook follows a consistent pattern. First, isolate which IP addresses are affected. If you send across multiple IPs, the problem may be confined to one that carries a particular mail stream (promotional vs. transactional, for instance). Second, check complaint rates and trap hits for that IP over the previous seven to 14 days. Third, review what changed in your sending during that window: new list segments, increased frequency, different content types, a new sending domain. Fourth, take corrective action on the root cause. Fifth, and this part requires patience, wait. Microsoft's reputation systems do not recover instantly. A clean sending period of seven to 21 days is typically needed before you see status improve from red back to green.
For senders managing large IP pools or multiple brands, the manual process of checking SNDS daily, correlating data with internal sending logs, and triaging reputation drops becomes operationally expensive. This is where algorithmic approaches to reputation management start to pay for themselves. Platforms that take this approach, such as Market Rithm's Deployer infrastructure with its patent-pending Adaptive Delivery technology, monitor reputation signals across mailbox providers in real time and adjust sending behavior automatically rather than relying on a human to catch a yellow status 24 hours after the fact. The difference between reactive and proactive reputation management is often the difference between a temporary dip and a weeks-long deliverability crisis.
There is a companion tool worth knowing about: Microsoft's Junk Mail Reporting Program (JMRP). Where SNDS gives you aggregate data, JMRP sends you individual feedback loop reports when a recipient marks your message as junk. You can sign up for JMRP through the same sender support portal. The combination of SNDS (macro view) and JMRP (individual complaint notifications) gives you a reasonably complete picture of your Microsoft deliverability health. Process JMRP complaints by automatically suppressing complainers from future sends. Every complaint you prevent reduces the complaint rate that SNDS reports, creating a positive feedback loop for your reputation.
Authentication also plays into how Microsoft evaluates your traffic, and SNDS data becomes much more interpretable when your authentication house is in order. At minimum, you need SPF and DKIM passing for every message, and a DMARC policy published at enforcement level (p=quarantine or p=reject). Microsoft has been increasingly aggressive about filtering unauthenticated mail, and starting in 2024, they began requiring DMARC compliance for high-volume senders to consumer mailboxes. If your SNDS data shows filtering but your complaint rates are low and you have no trap hits, the authentication stack is the first place to investigate.
The operators who get the most value from SNDS treat it as one input in a broader deliverability monitoring system, not as the whole picture. They combine SNDS with Google Postmaster Tools (for Gmail visibility), their own seed-list inbox placement testing, authentication monitoring, and internal engagement metrics. No single tool gives you complete deliverability intelligence, but SNDS is the only free, first-party source of truth for the Microsoft ecosystem. Ignoring it is like skipping your largest customer's satisfaction survey.
Set up SNDS this week if you have not already. Enroll in JMRP alongside it. Build a simple weekly review into your operations cadence, even if it is just a five-minute check of IP status and complaint rates. The senders who maintain strong inbox placement at Microsoft are not the ones with perfect sending records. They are the ones who catch problems early enough to fix them before the algorithms escalate filtering. Reputation management is a continuous practice, not a one-time configuration, and SNDS is the most accessible starting point you have for the mailbox provider that covers nearly a third of consumer email worldwide.
What is Microsoft SNDS and who should use it?
Microsoft Smart Network Data Services (SNDS) is a free portal that shows email senders how Microsoft's consumer mail network (Outlook.com, Hotmail, Live.com) perceives their sending IPs. Any organization sending more than a few thousand emails per day to Microsoft domains should be enrolled. It provides reputation status, spam trap hit data, complaint rates, and sample filtered messages.
How often should I check SNDS data?
For senders pushing more than 100,000 messages per day to Microsoft domains, daily checks are warranted. Lower-volume senders can review every two to three days. Because SNDS data is reported with a 24-hour delay, building a regular cadence is more effective than checking reactively after a problem surfaces in your open rates.
What does a red status on SNDS mean?
A red IP reputation status means Microsoft is aggressively filtering or blocking a significant portion of mail from that IP. It does not tell you why. You need to correlate the red status with complaint rate trends, trap hit data, and any recent changes in your sending program to identify the root cause. Recovery typically takes seven to 21 days of clean sending after the underlying issue is resolved.
Can I use SNDS if I send from shared IPs?
You can only enroll IPs you own or control. If your ESP uses shared IP pools, you will need to ask them to provide SNDS data or grant you access. If they cannot or will not, that is a meaningful gap in your deliverability visibility and worth factoring into your ESP evaluation.
What is the difference between SNDS and JMRP?
SNDS provides aggregate, IP-level data: overall reputation, complaint rates, and trap hits across your sending volume. JMRP (Junk Mail Reporting Program) sends you individual complaint notifications when a specific recipient marks your email as spam. Used together, they give you both the macro trend and the granular detail needed for effective reputation management at Microsoft.
