No More Spam... they said

Last week, Google announced its intention to fight spam harder than ever ... at least on their platform.

You can read the blog post here.

I’ll break the news down into 2 important questions:

How do they intend to do it exactly?

What does it mean for us cold emailers?

Let’s get started.

How do they intend to do it exactly?

In short, 4 ways:

1 - limit volume

2 - enforce technical setup

3 - require an unsubscribe link

4 - monitor spam rate on their system

1. Limit volume

Starting in Feb 2024, senders who send more than 5,000 emails per day to Gmail email accounts will need to comply with new rules described later in this article (source)

Technicality aside, you may be reading this and thinking that it won't apply to you.

After all 5k per day is plenty, but it’s not that crazy when you look at it closer.

The limit seems to indicate it's per workspace, not per email address or domain (the definition of the sender in their communication is unfortunately ambiguous, so it's best to assume the worst)

We need to consider sending follow-ups with our campaign (a fair assumption).

Assuming a 0% response rate and a 3-email step campaign, that means contacting 1,666 (= 5,000/3) new prospects per day.

Still high volume sending for sure, but imagine now that your market is really large and you have 10 sales reps, each running their own campaign.

Assuming the limit is per workspace (or even domain), not per email address.

Suddenly, each person can only contact 166 prospects per day.

It goes fast.

Although the vast majority of QuickMail accounts don't send over 5k emails per day, we do have a few clients sending more than 60k emails per day with success.

Now, we can mitigate those numbers as Google will only count emails sent to Gmail addresses (free and business).

So if contacting 5,000 email addresses means that you have to comply, if 10 of those are to Outlook email addresses, you would not need to comply.

If you are thinking of using inbox rotation to spread the load, this could work as long as your inboxes are spread across multiple Google Workspaces (or better yet different providers).

So there are ways around this limit.

2. Enforce technical setup

Google will prevent you to send emails or deliver them straight to the spam folder when it’s easy for hackers to spoof your address.

This may sound scary, but it's not. It just means you need to have SPF, DKIM, and DMARC setup.

We’ve added detections for proper SPF and DKIM configuration for years in QuickMail already and notify users when it’s not properly set, so this won't be a problem.

Pro tip: If you aren’t using QuickMail and want to check if your configuration is good, send an email to SpamTester.ai

DMARC is more annoying to set up (see this article), but since 'none' would work for Google, this is really just about having it defined in your domain configuration (DNS).

There is a series of rules to prevent people from building an SMTP server yourself and spamming people, but that's no real concern to most users on QuickMail using a famous ESP (like Google or Microsoft).

3. Require an unsubscribe link

Google does not actually say to add a link in the email body (although you can).

It asks that you provide such information in the header of the message (that is read by the Gmail server).

In QuickMail it's an option called "Add unsubscribe link in email header" in the campaign settings. It's been there for years and it's as simple as ticking a box to activate it.

If you are using another system, however, you'll need to check with them if they support this option.

4. Monitor spam rate on their system

When sending hundreds of emails per day, your domain will start collecting deliverability information.

You can then see this information in Postmaster Tool (tool from Google).

Google mentioned that the spam rate should be below 0.3% (ideally 0.1%).

Those numbers, however, are impossibly low for cold outreach.

This means that you should not get more than 3 emails flagged as spam for every 1,000 emails sent.

Let's run some maths.

Imagine that you have a bad campaign and 30% of prospects flag your email as spam.

This means that when sending 10 emails, 3 will flag you as spam. It's not that crazy since you are, after all, contacting total strangers.

Because you can only have 3 emails flagged as spammed for every 1,000 emails sent, you'll need to find other ways to send 990 emails to recipients you know won't flag you as spam (e.g. using an auto warmer like MailFlow) to be able to send only 10 cold emails…

Even with campaigns where "only" 10% of people flag the email as spam, you'll only be able to send 30 emails for the campaign and will need to use an auto warmer to send the 970 emails remaining.

So if you want to contact 30 people per day, your auto warmer should send 970 emails (and we aren't counting follow-ups).

That makes things very hard to scale.

Especially as Google email accounts can send up to 2,000 emails per day.

In practice, most auto warmers allow you to send up to 40-50 emails a day. Assuming a 10% spam rate, we would be able to send only 1 cold email outreach per day with those volumes.

Flagging an email as spam will equate to signing the death sentence of a sender.

It's unclear from their wording if they will keep information about senders from other providers (e.g. someone sending from Outlook) or only penalize senders using Google platform to send.

If only Google senders were going to be subject to this, we'd most likely see a flood of cold emailers moving to Microsoft or another email service provider as their sending platform.

What does it mean for us cold emailers?

Technical requirements are already well understood by cold emailers.

No one would launch a campaign without SPF and DKIM setup these days.

The clarity around unsubscribe links not impacting deliverability is welcomed.

And for those lagging, it's easy to turn on, it won't be any problem.

The spam rate to determine wanted emails however is another story.

If Google decides to enforce the spam rate as mentioned above, that could be the end of cold email as we know it for anyone sending more than 5k emails a day (at least from one sender).

This will set the bar incredibly high in terms of successful cold outreach campaigns.

It will be near impossible to maintain less than 3 spam reports per 1,000 emails sent and auto-warmer services will have only a limited use in this scenario.

Google didn't mention if they plan to do anything about lower sending volumes (less than 5k/day).

However, we can safely assume that Google will restrict the deliverability of email addresses getting a few too many spam reports at some point (it's already the case).

Newcomers may find it difficult to start with cold email in a less forgiving environment, leaving little room to learn.

Simply put, poorly executed campaigns may not have a second chance.

We will be forced to segment our list of prospects based on the provider used (e.g. contact only prospects who use Outlook).

Finally, we may see cold emailers move from using Google to using Outlook or other custom SMTP platforms.

Checking our QuickMail database gives us an interesting perspective, however.

Gmail addresses only account for about 25% of prospects (it's about the same for Outlook addresses).

So you could in theory contact 20k prospects with this distribution (since 25%, that is 5,000, would be Gmail) and still be OK, but better to be safe than sorry when calculating numbers.

Google has a history of announcing changes in October and implementing them in February of the following year.

This happened with Tensorflow, app auditing, auto-warmer ban…

We'll keep an eye on how it affects the sending of cold outreach and will let you know. Stay tuned :)