Inbox Testing: Find Out Where Your Email Actually Lands Before You Send

Most people measure email the wrong way. They look at a delivery rate of 99 per cent, see almost no bounces, and assume the campaign worked.

Delivered does not mean seen. A message can be accepted by Gmail, Outlook or a corporate mail server and still be filed straight into spam, or dropped into the Promotions tab where nobody looks. As far as your sending platform is concerned, that email was delivered. As far as your customer is concerned, it never arrived.

Inbox testing is how you close that gap. It tells you where your email is actually landing before you send it to your real list.

What inbox placement testing actually does

The idea is simple. Before your campaign goes out, you send the same message to a set of seed addresses sitting on the mailbox platforms your list actually uses. For Australian business sending, that means Google and Microsoft, across both personal and business accounts. Gmail and Google Workspace on one side, Outlook.com and Microsoft 365 on the other.

Those four cover the overwhelming majority of the addresses on a typical Australian list. Personal Gmail and Outlook.com accounts for consumer contacts, Google Workspace and Microsoft 365 for the business ones. Between them they set the filtering standard everyone else follows.

The test then checks where each copy ended up. Primary inbox, Promotions, Updates, spam folder, or blocked entirely. You get a placement breakdown per platform instead of one meaningless delivery number.

That distinction matters, because placement is rarely uniform. It is very common to see 90 per cent inbox on Microsoft 365 and 40 per cent on Gmail from the exact same send. It is just as common to see personal Gmail behave differently to Google Workspace, because business tenants layer their own filtering rules on top. An average figure hides all of it.

Why placement varies so much

Google and Microsoft score every message on a mix of signals, and they weight them differently. The big ones are:

Authentication. SPF, DKIM and DMARC need to be set up correctly and aligned to the domain you are sending from. Since the bulk sender requirements Google rolled out in 2024, and the equivalent tightening on the Microsoft side, DMARC is effectively mandatory for anyone sending at volume. Get this wrong and nothing else you do will save you.

Domain and IP reputation. Your sending history follows you. A domain that has been sending clean, engaged mail for two years is treated very differently to one that started sending last Tuesday.

Engagement. Opens, replies, deletes without reading, and spam complaints all feed back into how the next campaign is treated. Providers watch what recipients do, not what you claim.

Content and structure. Link shorteners, mismatched tracking domains, image only emails, spam trigger phrasing, and a broken or missing unsubscribe link all pull placement down.

List quality. Old or purchased data full of dead addresses and spam traps is the fastest way to wreck a domain that was previously fine.

An inbox test does not just give you the result. It shows you which of those levers is dragging you down.

What you do with the results

A test is only useful if it changes something. In practice, results usually point to one of a few fixes.

If placement is poor on both Google and Microsoft, the problem is almost always authentication or reputation. Check your DNS records first. Misaligned DKIM and a missing DMARC policy account for a huge share of the deliverability problems we see when auditing a new account.

If Google is weak but Microsoft is fine, it is usually engagement or content. Google leans harder on recipient behaviour than Microsoft does. Tighter segmentation, better subject lines, and cutting the disengaged portion of your list will move the number.

If Microsoft is weak but Google is fine, it tends to point the other way. Microsoft weights sending IP and domain reputation more heavily and is less forgiving of a cold or newly warmed domain, so the fix is usually slower volume ramping and cleaner list hygiene rather than a content rewrite.

If the business accounts are weak but the personal ones are fine, you are hitting tenant level filtering. Google Workspace and Microsoft 365 administrators can add their own rules on top of the platform defaults, and that sometimes needs a direct conversation with the recipient's IT team.

If you are landing in Promotions rather than spam, that is a softer problem. It is a content and formatting signal, not a reputation one. Fewer images, less commercial styling, and a more conversational structure often shifts it.

Testing is not a one off

Reputation moves. A single campaign to a stale list, a spike in complaints, or a new sending domain that has not been warmed can undo months of good behaviour.

The teams who stay in the inbox treat testing as routine. Run a placement test before any significant campaign, after any change to your sending domain or DNS, and whenever you start sending to a segment you have not touched in a while. If you are launching a new domain, pair testing with a proper warm up so volume ramps gradually instead of arriving all at once.

How it works in FlipSend

We built inbox placement testing into FlipSend because we got tired of clients finding out about a deliverability problem from their sales team rather than their reporting.

Inside the platform you can run a placement test on any campaign before you send it. We test across Google and Microsoft, personal and business, because that is where Australian mail actually lands. You get the breakdown for each, an authentication check covering SPF, DKIM and DMARC, and a content review flagging the elements likely to be hurting you. Domain warming runs alongside it, so a new sending domain can be built up on a schedule rather than dropped in cold.

Email validation sits in front of all of it. Lists are cleaned on import to strip dead addresses and traps, and we run a second automatic validation at the point of send to protect your domain reputation. That second check is on us, because it protects the platform as much as it protects you.

Reporting then separates genuine human opens and clicks from bot and privacy proxy activity, so the engagement numbers you are acting on are real ones. There is no point optimising against inflated open rates generated by security scanners.

FlipSend is Australian built and Australian hosted, with all processing and storage in AWS Sydney. For healthcare, government and other regulated senders, that means deliverability tooling that does not quietly move client data offshore.

The short version

If you are sending campaigns without testing placement, you are guessing. Delivery rates tell you the mail was accepted. Inbox testing tells you whether anyone will see it.

Run a test before your next campaign. If the result surprises you, that is exactly the point.

---

Want to see where your email is landing? Start a 14 day free trial of FlipSend at flipsidedigital.com.au, or get in touch for a deliverability audit of your current setup.