Every cold email platform comparison you'll find is written by the platform, or by an affiliate paid to recommend one. That makes third-party, methodology-disclosed testing rare enough to be worth a full write-up when it surfaces, especially when the gap it finds is large enough to change which platform you'd choose.
The test, and why the methodology matters
The test connected 24 mailboxes across six sending domains, warmed each domain for 21 days before sending a single cold email, then ran a 14-day campaign at 8,000 emails per day per platform. Inbox placement was verified daily using GlockApps seed tests, which place tracked messages across major inbox providers and report back exactly where each one landed: inbox, spam, promotions, or missing entirely. That is meaningfully different from a platform's own self-reported delivery rate, which typically only confirms the message left the platform's servers, not where the receiving mailbox actually filed it.
Twenty-one days of warmup before the test window started matters too, because it controls for the single biggest confound in any deliverability comparison: a cold, unwarmed domain will underperform on any platform, and a comparison that skips warmup discipline is really testing warmup speed, not inbox placement. Holding warmup constant across all three platforms is what makes this specific test worth citing instead of the dozens of comparison posts that skip the methodology section entirely.
The numbers
| PLATFORM | INBOX PLACEMENT | SPAM FOLDER | MISSING |
|---|---|---|---|
| Smartlead | 88% | 6% | 6% |
| Instantly | 81% | Not reported | Not reported |
| Lemlist | 74% | Not reported | Not reported |
The full breakdown of spam-folder and missing rates was published for Smartlead specifically: 6% landed in spam, 6% went missing entirely (neither delivered nor spam-foldered, meaning the receiving server accepted and then silently dropped it), alongside a sub-2.5% bounce rate and roughly 41% average open rate across the test window. Instantly and Lemlist's inbox-placement topline numbers came from the same test and methodology; their spam/missing splits were not broken out in the source review, which is worth flagging rather than papering over.
Inbox placement by platform, GlockApps-verified 14-day test at 8,000 emails/day, 24 mailboxes, 6 domains.
Fourteen percentage points, top to bottom, on identical sending volume with identical warmup. Run the math on a 10,000-contact list at that spread and you get roughly 1,400 more prospects seeing your message on the best-performing platform versus the worst, before a single word of copy or a single personalization variable changes anything. That is the size of gap that platform choice alone can produce, independent of list quality, offer, or sender reputation, which is precisely why it deserves more attention than most teams give it when picking a sending tool.
Why the gap exists at all
“A 7-to-14-point inbox placement gap is not a rounding error. At scale, it is the entire difference between a campaign that's profitable and one that's quietly burning the list it was supposed to convert.”
What the platform can't fix for you
None of this should read as "switch platforms and your deliverability problem is solved," because platform choice is one input among several, and it is not always the largest one. A perfectly deliverable platform still can't rescue a list built from stale, unverified contacts, and it can't override the sending thresholds Gmail and Microsoft now enforce regardless of which tool sits on top of them. We've covered the specific Gmail and Yahoo spam-complaint thresholds that apply no matter which platform sends the mail, and a platform with 88% inbox placement in a controlled test will still get throttled if the underlying sending pattern trips those thresholds.
Sender reputation is also cumulative and platform-agnostic in a way this kind of point-in-time test can't fully capture. A domain's sender score decays and recovers over weeks, not instantly on platform migration, which means a business currently struggling with deliverability on one platform won't see 88% placement the day after switching to a better-performing one. The reputation debt travels with the domain and the sending history, not just the tool.
| WHAT THE PLATFORM CONTROLS | WHAT IT DOESN'T |
|---|---|
| IP reputation infrastructure and isolation | Whether your list is verified and current |
| Automated warmup ramp quality | Whether your copy trips spam-filter language triggers |
| Authentication defaults and header hygiene | Gmail/Microsoft bulk-sender threshold compliance |
| Bounce and complaint handling automation | Existing domain reputation debt from before migration |
Reading the spam and missing split correctly
The Smartlead-specific breakdown, 88% inbox, 6% spam, 6% missing, is worth unpacking further because the two failure categories are not equally bad, and most teams treat them as interchangeable. A message that lands in spam is still, technically, deliverable: the recipient can find it, mark it as not spam, and some email clients even preview spam-folder subject lines in a way that occasionally earns a reply anyway. A missing message is a dead end. It was accepted by the receiving server and then silently dropped, with no folder, no trace, and no future chance of being seen. Two platforms could report the same 88% inbox rate with very different spam-versus-missing splits underneath it, and the one with a higher missing rate is the worse platform even though the headline number looks identical.
This is also where seed testing earns its keep over self-reported delivery metrics. A platform dashboard showing 98% delivered is answering a completely different question than "where did it land," and the gap between those two numbers is exactly where missing-rate problems hide. If your current platform can't show you a spam-versus-missing breakdown at all, that absence is itself useful information about how seriously the platform's own reporting takes the distinction.
The cost math at scale
Put real numbers against the 14-point gap to see why it matters more than it sounds like in percentage terms alone. A sender running 8,000 emails a day, the exact volume this test used, at 88% inbox placement gets roughly 7,040 messages into an inbox daily. The same volume at 74% gets roughly 5,920. That's a difference of over 1,100 inbox placements a day, every day, from platform choice alone, before any list-quality or copy variable enters the picture. Multiply that across a 90-day sending quarter and the gap is worth more than 100,000 additional inbox placements, which at even a modest 1% reply rate is over a thousand extra conversations a quarter that the lower-performing platform simply never generated a chance at.
That framing matters for how a team should actually evaluate platform pricing. A marginally more expensive platform that clears this kind of inbox-placement gap is very likely the cheaper option once you price in the pipeline value of the extra placements, not the more expensive one the sticker price suggests. Most procurement conversations about cold email tooling never run this math, because the deliverability gap is invisible until someone runs a seed test, and by then the contract is already signed.
Choosing based on your actual volume
The right read on a 14-point placement gap depends heavily on where you sit on the volume curve, and treating this test as a universal verdict misses that. A solo operator sending a few hundred emails a week will likely see smaller absolute gaps between platforms than this test found, because low volume means less exposure to the IP-pool and reputation-isolation mechanics that drive most of the difference at scale. A team sending 5,000-plus emails a day, where this test's 8,000/day figure sits, is exactly the volume tier where platform-level infrastructure differences compound fastest and matter most.
There's also a caution worth stating plainly: one 14-day window, run once, is a snapshot, not a permanent ranking. Platforms update their infrastructure, warmup algorithms, and IP pools continuously, and a gap this size measured in one testing window is not guaranteed to hold six months later in either direction. Treat this data as a strong signal to test your own sending patterns against your own domains before a platform migration, not as a permanent scoreboard to make a five-year decision from.
Do this next: if you're above the 1,000-emails-a-day tier and haven't run a seed test on your current platform in the last quarter, that's the actual first move, before evaluating whether to switch anywhere. If the number comes back materially below what this test found for your current platform, the gap is more likely sitting in your warmup discipline, list hygiene, or domain reputation debt than in the platform's baseline infrastructure. Our cold email team runs exactly this kind of seed-test audit before recommending any infrastructure change, because switching platforms is expensive and slow to reverse, and it is the wrong first move for a problem that often lives somewhere else entirely.
See where you are cited today
A free snapshot audit of your rankings and AI citations before we ever talk.
Tyler leads work at the intersection of SEO and generative engines at Something Inc., helping B2B brands get ranked and cited across every major AI engine.