Many organizations suffer the pain of emails ending up in spam folders or being throttled—often because sending practices are widely inconsistent with Outlook 365’s spam filters. At Warmy, our research team continues to do more by figuring out how to get more of your emails delivered, straight to your intended inbox. Our ongoing testing and analysis will enable you to avoid typical email-sending mistakes that make outreach wasteful for businesses.
It’s a fact: Every email counts, and even minor missteps in sending strategy can have a significant impact. That’s why we recently conducted an in-depth experiment comparing two distinct methods: grouped sending versus randomized sending.
What does sending strategy have to do with email deliverability?
It is not only the quality of the content that decides email deliverability. It’s also very much connected to the way you send your emails. How you distribute your messages can have a major impact on whether or not your email hits the inbox or the spam folder. Of course, it’s not the only factor, but it does have an impact.
Different sending strategies send different signals to email service providers:
- For instance, sending a bunch of emails at once, or all at the same time, could initiate “bulk spam activity” and trigger spam filters.
- Conversely, a more randomized sending approach could well resemble natural user behavior, possibly reducing the chances of being considered spam.
So, which approach is more effective? And how much does email-sending strategy really influence deliverability?
In our experiment, we put these two methods to the test, analyzing real-world results to determine which approach leads to higher inbox placement and better engagement. The findings might surprise you.
The experiment: grouped sending vs. randomized sending
What is group sending?
Group sending is a method where emails are dispatched in large batches over a short span of time.
However, sending a group of emails should also be done in moderation to avoid being flagged by email providers’ spam filters. Even elapsed time changes could be considered too close together after an unanticipated activation, which some providers (like O365) would monitor to detect spam-like activity.
In our experiment, we structured group sending to closely examine how these bursts of emails impact overall deliverability, ensuring that the method’s efficiency doesn’t come at the cost of inbox placement.
What is randomized sending?
On the other hand, randomized sending is sending emails at different intervals within a time frame.
Rather than sending a big batch at once, each email is sent out one at a time and in a seemingly random order. This approach aims to simulate organic, natural email send traffic which may reduce the probability of being flagged by spam filters.
By avoiding the concentrated burst characteristic of group sending, randomized sending aims to offer a smoother, more consistent deliverability profile.
Our experiment put this method to the test, comparing its effectiveness against group sending to see which approach better ensures that emails reach their intended inboxes.
Objective of the experiment
In our research, we dug deep into how these varying methods—grouped sending versus randomized sending—impact deliverability. Our goal was to identify which approach aligns better with the criteria used by providers like O365, helping organizations ensure their emails are not just sent, but actually seen.
Specifically, the experiment’s objectives were:
- To compare grouped sending and randomized sending strategies
- To measure impact using inbox placement and spam folder rates
Tools and methods employed
To rigorously test how different sending strategies impact email deliverability, our experiment was designed with clearly defined methodologies, focus groups, and performance metrics. Here’s a closer look at the setup:
Grouped Sending:
- 2 Focus Groups: 10 senders each (5 established, 5 new domains).
- Strategy: Send batches of 20, 100, and 300 emails daily over 3 weeks.
- Timing: Work hours (10 AM–6 PM) and “siesta time” (12 PM–2 PM).
Randomized Sending:
- 2 Focus Groups: 5 senders each (2 established, 3 new domains).
- Strategy: Sent emails in staggered batches:
- 40/week (8-hour spread), 120/week (24-hour spread), 300/week (randomized timing).
For both grouped and randomized sending experiments, we used both Gsuite and Custom SMTP setups to simulate typical sending environments. All emails were directed to Microsoft 365 accounts, ensuring a consistent testing environment across both strategies.
To accurately assess deliverability and inbox placement, we used Warmy’s in-house tools alongside Microsoft SNDS. Then, we tracked the following metrics:
- Inbox vs. Spam Rates: To determine the effectiveness of each sending strategy.
- Bounce Rates: To capture instances of emails being blocked or delayed.
The results: So how do sending methods impact deliverability?
Imagine you’re hosting a grand dinner party. You could either invite everyone all at once—risking chaos at the door—or you could welcome guests gradually, ensuring each one gets the attention they deserve. Our experiment on email deliverability revealed a similar story between two sending strategies.
Grouped sending: The high-volume gamble
At the start of our experiment, sending emails in large bursts actually showed promise. In Week 1, small batches achieved a decent 75% inbox placement. However, as the volume ramped up by Week 3, the strategy faltered dramatically: more emails were landing in spam (55%) instead of inboxes (35%).
Randomized sending: The steady performer
In contrast, randomized sending told a very different story. This approach consistently delivered strong results—regardless of whether we sent 40, 120, or even 300 emails per week.
Randomized sending showed the following results:
- Consistent deliverability: Inbox rates remained high (80–90%) even at 300 emails/week.
- Low spam rates: Spam placement stayed below 15%.
Minimal Loss: Only 2–5% of emails were not received.
Why grouped sending fails
Analyzing the trends and data from the experiment, it became clear why sending emails in large, grouped batches consistently failed over time. The key issues that emerged include:
Spam filters
Most email providers (notably Outlook (O365)) have advanced spam detection systems that can recognize when emails are suspicious. When thousands of emails are sent at a time from the same domain, Outlook identifies this as mass marketing or spam.
So what happened during the experiment was Outlook flagged bulk sends as suspicious, which led to higher spam placement (up to 55%). Additionally, the more frequently a domain engages in grouped sending, the more Outlook reinforces its classification as a potential spam sender.
Throttling
High volume triggers temporary blocks or delays, resulting in 10% of emails not being received.
Many email servers, including O365, use throttling mechanisms to prevent abuse and protect users from spam.
When a server detects a sudden surge in outbound emails, it may impose temporary restrictions, including email delays. Instead of immediate delivery, some emails may get queued, leading to significant delays in inbox placement. The receiving server may also respond with temporary errors and if high-volume sending continues, the sender’s IP or domain reputation may suffer long-term consequences.
Reputation risk
Beyond immediate deliverability issues, bulk sending can hurt the sender’s reputation in the long run. If a domain is frequently marked as spam, its emails will be more likely to land in spam folders—even when sending to new recipients. A damaged reputation also means that even well-crafted, legitimate emails may struggle to land in inboxes.
Why randomized sending works
Meanwhile, here is our analysis on why randomized sending resulted in low spam rates and consistent deliverability—even while sending up to 300 emails in a week.
Trust building
Email providers have built-in algorithms that assess email-sending behavior. Gradual sending aligns with O365’s expectations of “normal” email traffic. Though this was automated, it simulated human behavior, which made it more credible to O365. In this experiment, randomizing email dispatch times led to:
- Less suspicion from spam filters: Since the sending pattern resembled that of a typical user rather than a bulk sender, Outlook’s filtering systems were less likely to intervene.
- Improved domain reputation: The lack of bulk activity helped maintain a positive sender score, ensuring future emails were trusted by the receiving servers.
Engagement wins
The time an email arrives in someone’s inbox can significantly impact whether they open it. Recipients are more likely to open emails that arrive at staggered times. Bulk emails often get buried under other promotional emails sent at the same time. When emails arrive at varied intervals, recipients are more likely to see and interact with them, leading to higher open and response rates.
Minimal loss
A key metric in the experiment was email acceptance rate, and randomized sending showed a clear advantage. Only 2–5% of emails were not received, compared to the 10%+ failure rate observed with grouped sending. Since emails were sent in a steady flow instead of overwhelming the server, Outlook accepted them more consistently.
The conclusion: key findings from the experiment
The findings from this research make one thing clear: randomized sending significantly outperforms bulk email sending when targeting Office 365 recipients. By understanding how Outlook’s filtering systems interpret email activity, we can see why a strategic approach to email distribution is essential for long-term deliverability and sender reputation.
Inbox placement: more emails landing in the right place
Randomized sending achieved an impressive 80–90% inbox placement rate, compared to bulk sending’s 35–75%. That means more emails land in the recipient’s main inbox, where they have a better chance of being opened and acted on.
Why this matters: The greater the number of emails that land in the primary inbox, the more likely they are to be seen, opened and taken action on—improving engagement and conversion rates.
Spam rates: reducing risks of being flagged
Bulk sending saw up to 55% of emails flagged as spam, and randomized sending kept spam rates between 7–15%. A major factor here is the reputation of the sender and avoiding blacklists.
Why this matters: That’s not just a challenge for that particular email campaign; every email flagged as spam has a cumulative effect. Once flagged, it’s much more likely that subsequent emails from the same sender will be filtered out as spam as well, making it even harder to reach those intended recipients.
Delivery reliability
With randomized sending, only 2–5% of randomized emails were undelivered, versus 5–10% in bulk sending. This ensures your messages aren’t lost to throttling or blocking.
Why this matters: Lost emails result in lost effort, lower outreach success, and missed chances to reach recipients
Recap: why randomized sending works
Randomized sending mimics natural email behavior, which aligns with O365’s spam filters and reputation algorithms. By sending emails in smaller, staggered batches, you avoid triggering red flags associated with high-volume campaigns. This approach not only improves deliverability but also enhances recipient engagement, as emails arrive at more natural intervals.
Recap: why bulk sending fails
Bulk sending, while efficient for large campaigns, often backfires with O365. The platform’s filters interpret sudden spikes in email volume as suspicious activity, leading to higher spam placement, throttling, and even temporary blocks. Over time, this can damage your sender’s reputation, making it harder to reach inboxes in the future.
The verdict: which strategy wins?
For businesses aiming to maximize email deliverability and engagement with O365 recipients, randomized sending is the superior strategy. It’s a minor change that has big effects: more emails in inboxes, fewer in spam boxes, and better performance all around.
Instead of bulk sending (which poses the risk of spam filters, blacklisting, and delivery failures), controlled, strategic email distribution guarantees inbox placement success, lower spam levels, fewer delivery failures, and a cleaner sender reputation.
With simple modifications in how emails are sent—the business can see improved email marketing effectiveness; providing a boost in open rates, response rates, and potential improvements in the long run.
From experiment to execution: Warmy’s role in email success
The experiment clearly showed that randomized sending triumphs over bulk, grouped sending. But what does this mean for your email campaigns? Getting the email sending strategy down pat is great, but it’s just one factor that affects your overall deliverability—and here’s where Warmy.io comes in.
Warmy’s platform is built on the very principles our research confirmed: consistency, natural email flow, and smart timing. Here’s a quick overview of Warmy’s capabilities:
- By harnessing advanced deliverability tools and real-time analytics, Warmy automates the process of “warming up” email accounts by mimicking natural email activity.
- It can handle up to 5,000 emails per day while simulating human-like interactions like sending, replying, and marking emails as important.
- Additionally, Warmy’s new Domain Health hub gives users the ability to monitor deliverability at the domain level. This includes spam rates, inbox placement, and deliverability trends on a weekly & monthly basis.
For a more comprehensive breakdown of the numbers this experiment churned out, you can download the full report here.
