Choosing an email tool: what decides whether the mail arrives

Every comparison of email marketing platforms is a table of automation features and a price per contact. Both are real. Neither answers the question you are actually buying the tool to solve, which is whether the message lands in an inbox.

Since February 2024, that question is settled mostly outside the platform. Gmail and Yahoo published requirements for bulk senders and began enforcing them, and the requirements are about your domain and your recipients' behaviour — not about which vendor sends the message.

If you get that part wrong, the best platform on the market delivers your campaign to spam. If you get it right, a modest platform delivers it to the inbox.

What Gmail now requires, in its own words

Google defines a bulk sender as one transmitting more than 5,000 messages per day to Gmail accounts. For those senders, its published guidelines require:

• SPF and DKIM authentication set up for the sending domain.
• DMARC for the sending domain. Notably, the enforcement policy "can be set to none" — the record must exist, but it does not have to reject anything.
• A TLS connection for transmitting mail.
• Valid forward and reverse DNS records for sending domains or IPs.
• One-click unsubscribe on marketing and subscribed messages, plus a clearly visible unsubscribe link in the body.

And then the requirement that behaves differently from all the others:

Keep spam rates reported in Postmaster Tools below 0.10%, and avoid ever reaching a spam rate of 0.30% or higher.

The 0.30% is smaller than it sounds

Three tenths of one percent is three complaints per thousand delivered messages. On a list of 10,000, thirty people pressing "report spam" puts you at the ceiling.

That is not a large number of annoyed people. It is a normal number of annoyed people on a list that was built carelessly — bought, scraped, imported from an old CRM, or grown through a checkbox that was pre-ticked.

Two consequences follow, and both are uncomfortable for anyone shopping by feature list:

The metric is a rate, so a small list gives you no room. Ten complaints on 2,000 messages is 0.5%. The same ten complaints on 50,000 is 0.02%. Early sends are the most dangerous ones, and that is exactly when people are testing a new platform.

Nothing your vendor does can fix a list problem. Deliverability features, warm-up tools and dedicated IPs change how mail is transmitted. They do not change whether recipients wanted it.

The setup detail we nearly got wrong on our own domain

We authenticated our own domain, keenvale.com, on Google Workspace, and one step is easy to skip because everything appears to work without it.

By default, mail sent through Google Workspace is DKIM-signed with a shared Google key on a gappssmtp.com subdomain. Mail flows. Tests pass. It looks configured.

But that signature is not yours. Turning on DKIM for your own domain is a separate, explicit action: you generate a key in the Admin console, publish it as a DNS TXT record, and then switch signing on.

How to check which one you have, without trusting any dashboard: send a message to an address you control, open the raw headers, and find the DKIM-Signature. Look at the d= value.

• d=yourdomain.com with s=google — your own key is signing.
• d=something.gappssmtp.com — you are on the shared default.

We also kept our SPF record to a single include and set DMARC to p=none to begin with, which is what Google's own guidance permits while you observe reports before tightening.

Why this matters for a tool comparison: none of it is a feature of an email marketing platform. It is DNS, and it is the same work whichever vendor you pick. A platform that walks you through it clearly is worth more than a platform with twice the automation blocks and a vague setup wizard.

What the platform genuinely does affect

Having said the tool matters less than the marketing implies, it does not matter zero. Four things are real differences:

Whether you send from a shared or dedicated IP. On shared infrastructure your reputation is partly other customers' behaviour. On dedicated you own it entirely, which is better once you have volume and worse before you do, because a cold dedicated IP has no reputation at all.

Whether the one-click unsubscribe header is implemented properly. The requirement is a specific technical header defined in RFC 8058, not merely a link in the footer. Most established platforms handle it. Verify rather than assume, because a visible link alone does not satisfy the requirement.

Whether you can see your own complaint rate. Google Postmaster Tools reports spam rate by domain, and it is free. A platform that surfaces bounce and complaint data next to your campaign report shortens the loop; one that shows only opens and clicks lets a problem grow quietly.

Whether list hygiene is possible or painful. Suppression of hard bounces, automatic removal of long-term non-openers, and easy segmentation are the levers that hold the complaint rate down. These are the features worth comparing, and they are usually buried under the automation-builder screenshots.

How to compare platforms on the axis that matters

1. Does it publish clear, specific DNS setup instructions for your own domain, including DKIM?
2. Does it show bounce rate and complaint rate per campaign, not just opens?
3. Does it implement one-click unsubscribe as a header, and will support confirm that in writing?
4. Can you suppress and segment without exporting to a spreadsheet?
5. What does it cost at the list size you will have in a year, not today?

That last one is where price comparisons mislead most. Nearly every platform in this category prices by contact count, and the curve steepens where a growing list actually lands.

What we have not tested

We have authenticated a domain and read the current published requirements. We have not run bulk campaigns through competing platforms, so we cannot tell you whose delivery rates are better in practice — and we are sceptical of anyone who claims to, because deliverability depends on the sender's list far more than on the vendor, which makes clean comparison nearly impossible.

We have not measured our own spam complaint rate at volume, because we do not yet send at volume.

Requirements in this area changed in 2024 and have been tightened since. The numbers quoted here are from Google's published guidelines as of August 2026, and are worth rechecking against the source before you rely on them.