How to Check Whether Your Signup and Reset Emails Reach Customers
A newsletter landing in your inbox tells you nothing about whether your password-reset email works. Your product might send those messages through different services.
For a small SaaS, a useful email deliverability test starts inside the product. Create an account. Request a reset. Subscribe to your newsletter. Then check what actually arrives, where it lands, and whether you can use it.
Test the emails your customers depend on, using the same actions your customers take.
What should an email deliverability test check?
An email deliverability test should check whether a real message arrives, where it lands, what its headers report, and whether its links work. For your SaaS, that means testing the messages your product actually sends, rather than relying on a preview from your marketing tool.
Start with the paths named in this checklist:
- A signup email triggered by creating an account.
- A password-reset email triggered from the login screen.
- A newsletter email triggered through your subscription process.
Use accounts you control at Gmail and Outlook. These give you specific places to inspect the result. They don't prove what every customer will receive, so record each result as an observation, not a universal pass.
For every message, answer a few plain questions. Did it arrive? Did you find it in the inbox or spam folder? How long did you wait? Do the headers show passing authentication results? Can you complete the action the email promises?
Keep those answers separate. An email can reach the inbox but contain an unusable reset link. A newsletter can look fine while a signup email takes a different route through another sending service.
Don't compress those different outcomes into a single reassuring score. Write down what happened to each message.
A useful result sounds like this: “The password-reset message reached the Outlook inbox, but the link failed when opened.” That's a clear problem you can investigate. “Email passed” leaves too much hidden.
Decide what counts as success before you start. For a reset email, success means receiving the message and setting a new password. Seeing the subject line isn't enough.
Trigger each email through the product
Creating a real account tests the signup path your customer uses. Sending yourself a sample message from an email dashboard doesn't exercise that same sequence.
Open your product as a customer would. Use a test account you control, and avoid shortcuts that skip the action you're trying to check. If you're testing signup, start at signup. If you're testing a reset, start at the login screen.
Run this sequence for your Gmail account, then repeat it for Outlook:
- Create a fresh account and inspect any signup or verification message.
- Request a password reset and keep the resulting message separate.
- Subscribe through your newsletter form and inspect the email that follows.
If your newsletter requires confirmation, complete that step too. If it doesn't, test the path you've actually built. The goal isn't to invent extra emails. It's to check the promises your product already makes.
Record when you took each action. Otherwise, an older reset email can be mistaken for the one you just requested. Keep enough context to match the message to its trigger without storing passwords or live reset links in your notes.
Use a clean test account where possible. That makes it easier to tell whether the latest action worked, rather than guessing from an account with a long history of test messages.
You don't need a complicated test setup to start. You need a repeatable path from a customer action to a message you can inspect. Save that path as a short procedure so the next check doesn't depend on your memory.
A newsletter pass doesn't clear your password-reset emails
Your newsletter and password-reset emails can use different sending services. A successful newsletter test therefore doesn't establish that your product's other emails work.
Treat the sending service as something to record for each message, not something to assume across your whole product. Your notes should connect the customer action to the email it produced.
Build a small map with these fields:
- Message: signup, password reset, or newsletter.
- Trigger: the exact action you took in the product.
- Visible sender: the name and address the recipient sees.
- Sending service: the service your product uses for that message.
- Recipient: the Gmail or Outlook account used for the check.
Confirm the sending service in your own settings rather than guessing from the visible sender. If you can't identify it, mark it as unknown and investigate before treating the record as complete.
Then compare the results without merging them. A newsletter reaching Gmail is one observation. A reset email reaching Outlook is another. Neither should quietly stand in for a test you haven't run.
For a tiny SaaS, this keeps the work manageable. Instead of vaguely investigating “email,” you can name the failing path: password reset, sent through a particular service, to a particular mailbox.
It also makes changes easier to check. If you replace the service that sends reset emails, rerun that path. Don't point at a successful newsletter campaign as evidence that the new setup works.
Your customer won't know which service sent the message. You need that detail so you can find and fix the right thing.
What should you inspect in email headers?
Inspect each test message's headers for SPF, DKIM, and DMARC results. Record what the receiving mailbox reports instead of assuming that a setting in your sending tool proves the message passed.
Open the message's full headers or original-message view in Gmail and Outlook. The labels and controls may differ, so use the mailbox's current help instructions if you can't find them.
For each message, capture:
- The reported SPF result.
- The reported DKIM result.
- The reported DMARC result.
- Any missing result or error you need to investigate.
Keep the results attached to that specific email. If your newsletter and reset messages use different sending services, inspect both. Copying the newsletter's results into the reset row defeats the test.
Don't turn a missing or unclear result into a pass. Write down what you can see, keep the relevant header details privately, and investigate the gap. An honest “not checked” is more useful than a green box based on a guess.
Authentication is one part of this checklist. Continue checking placement, arrival time, and links even when the headers look right. A passing header result doesn't show whether the customer can finish resetting a password.
When you need to change sender settings, check your mailbox providers' current official requirements directly. Don't treat an old DNS tutorial as the final authority. This checklist doesn't specify current provider thresholds or certify compliance with their rules.
After a change, send a fresh message through the real product path and inspect its headers again. Keep the earlier result so you can compare what changed, rather than relying on the settings screen alone.
Check arrival time and finish the action
A password-reset test isn't complete until you use the email to set a new password. Finding the message in the inbox only answers part of the question.
Record the time you request the reset and the time you first observe the message. If you're checking manually, call that an observed arrival time. Don't present a rough check as a precise measurement.
Inspect the inbox and spam folder in each test mailbox. Write down where you found the message before moving it. If you don't find it, record “not found during this check,” along with when you looked. That keeps an observation from becoming a claim you can't support.
Then test the message as a customer:
- Open the email and check that you recognise the sender.
- Click the intended action link.
- Complete the reset or verification flow.
- Confirm that the product accepts the completed action.
For a password reset, try signing in with the new password. For signup verification, check that the account reaches the expected state. For a newsletter, check that the intended links open the right pages.
A reset email can arrive but still leave the customer stuck if its token has expired. Record a failed link separately from a missing message. They need different investigations, even though the customer experiences both as “I can't get in.”
The useful finish line is a completed customer action, not a message sitting in an inbox.
Don't quietly work around a broken step during testing. If you need an admin tool to finish, write that down as a failure of the customer path.
How do you record results without building another dashboard?
A short checklist can capture the sender, service, authentication results, placement, timing, and link outcome for each test email. A document or spreadsheet is enough to start.
Give every message its own row. Keep Gmail and Outlook results separate, because you're recording what happened in each mailbox. Avoid a single checkbox labelled “deliverability” that hides the actual evidence.
Use a template like this:
Field What to record Test date When you ran the check Customer action Signup, password reset, or newsletter subscription Recipient Which test mailbox you used Sender The name and address shown in the email Sending service The service responsible for that message Authentication SPF, DKIM, and DMARC results from the headers Placement Inbox, spam, or not found during the check Timing Trigger time and observed arrival time Link result What happened when you used the email Next action The specific issue to fix or investigate
Write failures in concrete language. “Reset link opened an expired-link page” gives you something to reproduce. “Email issue” doesn't.
Keep sensitive details out of shared notes. You can record that a reset link failed without pasting the live link or token into a document. The checklist should help you repeat the test, not collect account access details.
If you can't explain a result, leave the uncertainty visible. “Sending service unconfirmed” creates a clear follow-up. Guessing a service name makes the record look finished while leaving the question unanswered.
Save the procedure alongside the results. A future check should use the same customer actions, not whatever shortcut feels convenient that day.
When should you repeat the test?
Repeat the affected email tests after changing the sender, sending service, authentication settings, or customer flow. A saved result describes the setup you tested, not every setup you might run later.
Use the size of the change to choose the scope. If you change the reset-email path, test password resets in both Gmail and Outlook. If you change settings shared across your emails, rerun the full checklist.
Useful reasons to repeat the check include:
- Switching the service that sends product emails.
- Changing the sender address or related authentication settings.
- Editing signup, reset, or subscription behaviour.
- Changing a message's links or the pages they open.
- Investigating a customer report that an email didn't arrive or work.
Keep the previous record. Comparing the same path before and after a change gives you a clearer starting point than trying to reconstruct an old setup from memory.
When a customer reports trouble, don't dismiss it because your own test passed. Your test accounts show what happened in those accounts. Ask which action they took and what happened next, without requesting passwords or reset tokens.
Likewise, don't promise that a successful check guarantees every future inbox placement. The useful claim is narrower: you tested these customer actions, in these mailboxes, and recorded these results.
That narrower claim is still worth having. It replaces “the email tool says it's fine” with evidence from your actual product.
Before your next release, create a test account, request a password reset, and subscribe through your newsletter form. Run those actions with Gmail and Outlook accounts you control. Save the results, fix the failed step, and repeat it until you can complete the customer action.