When Outlook inboxes outperform Google Workspace for cold email. Three ICPs where I switched and what GlockApps showed
outlook_test_nina · 2026-07-07 · 1,160 views
Everything you read says Google Workspace is the gold standard for cold email infrastructure. I believed it until I ran a side-by-side test across three ICP campaigns and Outlook won on two of them.
Here is the honest breakdown.
The test setup.
Same copy. Same list segments. Same sending platform (Smartlead). One batch from PuzzleInbox Google Workspace inboxes, one batch from PuzzleInbox Outlook inboxes. GlockApps placement tests before and during each campaign.
ICP 1: Mid-market finance and accounting firms.
Google Workspace inbox placement: 79 percent. Outlook inbox placement: 91 percent. Reply rate from Outlook: 3.8 percent. From Google Workspace: 2.4 percent. Finance companies tend to run on Microsoft 365 internally. Emails from Outlook accounts may pattern-match better against Microsoft's spam filters. I can't prove causation but the result was consistent across three separate test runs on this ICP.
ICP 2: Healthcare practices and medical offices.
Same pattern. Outlook outperformed Google Workspace on inbox placement by 14 percentage points. Most healthcare organizations run on Microsoft 365 internally.
ICP 3: SaaS and tech companies.
Google Workspace won here. Inbox placement 88 percent versus 76 percent for Outlook. Tech ICPs skew toward Google Workspace internally and there may be preferential treatment from Google servers for same-ecosystem sends.
What I took from this.
Match your sending infrastructure to the technology stack your ICP runs internally. If your ICP is predominantly on Microsoft 365, test Outlook inboxes first. If your ICP is predominantly on Google Workspace, start with Google. PuzzleInbox offers both options, which is exactly why I use them for multi-ICP campaigns. For finance, healthcare, or government, I now default to Outlook. For tech, I default to Google Workspace. Run GlockApps before every new campaign segment. Don't assume your current infrastructure is optimal for a different ICP without testing it first.