RE:markable is the weekly email about emails. Dropping the latest email marketing news, updates, insights, free resources, upcoming masterclasses, webinars, and of course, a little inbox mischief.
Email deliverability often feels like some cryptic black box that marketers have to bow to.
One day, your emails are in the inbox and the next? Lost in the digital void of spam.
But here’s the truth most people don’t want to tell you: deliverability isn’t mysterious (or difficult to track, monitor and look after).
It’s not about praying to the gods of SPF, DKIM, and DMARC (although this is really important). And no, you don’t need to be an email engineer to land in the inbox.
Deliverability is mostly common sense. About 80% of it comes down to how your audience behaves, not the tech stack behind the send - and this is really good thing.
In this blog, we’re pulling back the curtain. You’ll learn how Gmail, Yahoo, and Outlook actually filter your emails, and how to turn all that noise into a practical, human-first strategy.
Spam complaints, opens, deletes and dragging emails to folders…these actions train the filters more than any DNS record ever could.
PSA: Email authentication is a must, but Outlook has it's own rules on this and Gmail has other rules.
Every individual recipient's behaviour contributes to your senders reputation, especially on Gmail and Yahoo, where deliverability is highly personalised and based on individual user interaction history. So a handful of disengaged users won’t directly penalise your engaged ones — but they do dilute your overall sender reputation. And if you keep sending to disengaged users over time, that pattern can get flagged as spammy behaviour, reducing your inbox placement across the board.
Yahoo and Gmail both use machine learning models that learn per recipient, but they also use aggregated signals to assess if a sender is “likely to be wanted” inbox-wide.
How Gmail knows your post-send engagement
Gmail is not just an inbox. It’s part of the broader Google ecosystem and that gives it visibility into user behaviour well beyond the email itself.
Here’s how it works:
1. Web tracking via Chrome, Android, and Google Search
2. Gmail user signals = deliverability gold
Gmail watches what users do, not just what senders send.…are all positive signals that your emails are valuable.
3. Behaviour after the send
Gmail infers relevance from downstream activity. For example:
If a recipient receives your email, ignores it, but visits your site via another route, that still counts.
Clickless conversions (i.e. they don’t click the email, but take action anyway) still improve your reputation.
4. Machine learning model per user
Gmail treats every recipient individually. So your sender reputation isn’t global; it’s personal per subscriber. A+ with one user, C- with another.
So what does that mean for you?
5. Email authentication
If your SPF, DKIM, or DMARC setup fails or is missing, Gmail immediately treats your message as suspicious. Even if the content looks clean, unauthenticated messages are far more likely to land in Spam, especially for new or low-reputation senders. Gmail uses authentication to confirm you're really you, fail that, and you're flagged fast.
Yahoo Mail might not get the same spotlight as Gmail, but its filtering systems are just as critical, and just as complex. While it leans on more traditional spam filtering models, Yahoo’s rules are no less strict, especially for senders with inconsistent behaviour or shaky domain reputation.
Your domain and IP reputation matter (a lot)
Yahoo evaluates sender reputation at both the IP and domain level, using historical data to determine how trustworthy your emails are. Factors that impact this include:
Frequency and volume of sending
Bounce rates and invalid addresses
Complaint rates (especially "Mark as Spam")
Spam trap hits
Consistency and cadence
If you send erratically, ramp up volumes too quickly, or switch IPs/domains without warming them up, Yahoo is likely to treat you cautiously, or block you outright.
Greylisting: The silent gatekeeper
Greylisting is Yahoo's way of saying, "Let’s wait and see." It’s a delay tactic, not an outright block, where Yahoo temporarily rejects messages from unknown or inconsistent senders. It does this to gauge:
If your server retries delivery (legitimate servers will)
How your sending behaviour develops over time
Whether user engagement follows (positive or negative)
What it looks like:
Delays in delivery (emails appear hours late)
No bounce message, but no immediate inbox placement either
Often affects new senders, new domains, or senders with volatile sending patterns
How to get through it:
Maintain consistent sending volumes
Use warmed-up, reputable IPs and domains
Don’t suddenly start blasting large lists from a new setup
Monitor Postmaster Tools and soft bounces
User feedback is everything
Like Gmail, Yahoo trains its filters based on what users do:
Marking your emails as spam or junk = 🚩 red flag
Creating filters to move your emails = a sign you're not relevant
Ignoring or deleting emails consistently = weak engagement signals
To strengthen inbox placement:
Ask users to add you to their contact list or safe sender list
Use clear unsubscribe options (hard-to-leave = fast way to complaints)
Send based on behaviour: segment active users, suppress unengaged ones over time
URLs can get you flagged
Yahoo's filters heavily scrutinise your email links. Using shady or mismatched URLs is one of the fastest ways to land in the spam folder.
Common red flags:
URL shorteners (bit.ly, tinyurl) without proper configuration
Domains with a history of abuse or flagged by blocklists
Misalignment between visible text and actual destination URL
Broken, expired, or dynamic links from unfamiliar domains
Best practice:
Use branded tracking domains
Ensure domain reputation is clean (check tools like Spamhaus, SURBL)
Keep a consistent domain identity across all email assets
Also, of course email authentication...
Yahoo’s spam filters treat authentication as a baseline trust signal. If SPF/DKIM/DMARC records are misconfigured or missing, your domain reputation suffers, and you're more likely to be greylisted, delayed, or dumped into Junk. It's one of the first things Yahoo checks before even looking at user behaviour.
Microsoft’s email filtering stack is powerful, enterprise-grade, and very rules-based, but it’s also incredibly user-aware. Outlook (via Exchange Online Protection, or EOP) relies on a combination of machine learning, content scanning, reputation analysis, and Spam Confidence Level (SCL) scoring to decide where your email lands.
It can be frustratingly rigid if misunderstood. So here’s the play-by-play:
Every email that passes through Microsoft’s email ecosystem is assigned an SCL score (0–9). This score determines whether your message goes to the Inbox, Junk folder, or gets outright blocked.
Key SCL thresholds:
SCL -1: Always deliver to inbox. Reserved for trusted senders like those on the safe sender list or internal domains.
SCL 0–1: Delivered to inbox. Considered not spam.
SCL 5–6: Likely spam. Routed to Junk Email folder.
SCL 9: High confidence spam. Often quarantined or rejected.
Microsoft SCL documentation (official)
EOP filtering technologies overview
SmartScreen is Microsoft's proprietary machine learning filter that assesses:
Content patterns (including word use, formatting, headers, metadata)
Sending behaviour (volume changes, list hygiene, historical complaints)
Reputation signals from your IP/domain
Global user interaction data across the Microsoft ecosystem (Outlook, Hotmail, Office 365)
Over time, it learns from feedback and adjusts filter thresholds based on whether people are:
Opening or replying
Deleting without reading
Marking as spam
Marking not spam (important signal)
SmartScreen doesn’t just judge your message once. It keeps adapting, which means if your emails consistently get ignored or flagged, you’ll sink - very fast.
SmartScreen Filter - Microsoft
One of Outlook’s most unique features is how much control it gives users over their filtering experience.
Users can:
Mark a sender or domain as safe (SCL score is overridden)
Use custom mail rules to route messages
Add to Blocked Senders (message goes straight to Junk regardless of SCL)
Right-click and mark a message as “Not Junk” (feeds SmartScreen training and bypasses future filtering)
Safe and blocked senders in Outlook
Authentication failures (SPF, DKIM, DMARC) = Instant suspicion
No engagement across multiple sends = Drops sender trust
Inconsistent “From” headers or reply addresses = Red flag
Use of dynamic or suspicious URLs = Common spam signal
Bulk sending patterns without segmentation = Lowered score
Inbox placement isn’t something you brute-force your way into. You’re not getting preferential treatment from Gmail or Outlook just because your SPF records are spotless or your DMARC policy sounds fancy.
You get into the inbox by sending better email. Email that people actually want. Email that lands when it makes sense. Email that respects the person reading it.
That means:
Relevant content based on real behaviour and lifecycle stage.
Timely sends that fit your customer’s journey, not your internal calendar.
Segmented lists that filter out the ghosts and serve the gold.
Consistent messaging that builds brand memory and trust.
Deliverability isn’t about tech perfection; it’s really about relationship consistency. And nothing tanks a relationship like shouting into someone’s inbox when they didn’t ask for it (or who has shown you they don't care after LONG time).
If your weekly KPI review still starts with "open rates," and "Click rates" we need to talk.
Open rates are shaky, click rates are shallow, and deliverability is bigger than any one campaign. Great email marketers have graduated from chasing vanity metrics to measuring real business impact.
So what should you actually be tracking?
Signals of impact:
Direct and organic traffic increases — people are Googling you or typing your URL after reading your emails.
Reply sentiment — are they writing back with enthusiasm or confusion?
Pipeline momentum — are email nurtures moving leads to sales conversations?
Brand search volume — are more people looking for you after campaigns land?
Retention curve stability — are subscribers sticking longer and engaging more?
If you paused email for six weeks and your funnel slows down, that’s not a problem, it’s proof. That’s the signal. That’s impact. That means your emails are doing their job in the background, even if there’s no UTM tag to track it.
I’m Beth O’Malley, email consultant, strategist, and the person marketers and businesses call when they’re ready to transform how email works for their company.
If your email strategy is more “send and hope” than structured and strategic, let’s change that. Whether you need a total reset, a smarter approach, or just want someone to tell you what’s worth fixing — I’ve got you.
Let’s have a conversation about working together.
RE:markable is the weekly email about emails. Dropping the latest email marketing news, updates, insights, free resources, upcoming masterclasses, webinars, and of course, a little inbox mischief.