Yes, Google requires opt-in

Every now and then, someone shows up in an email marketing or deliverability forum or chat room, trying to make the case that spam is somehow allowed. The argument usually goes like this: "The mailbox providers don't actually require permission, so it's fine if I just send email to whoever I want." After all, other people are sending spam. I'm receiving spam, so spam must be a normalized and expected practice, right?

Nope, not right. It is, in fact, wrong.

Spam slipping through filters isn't proof of tacit approval. It's just proof that filtering is imperfect. At very low volumes, there sometimes aren't enough data points to be able to tell with certainty that an email sender is a spammer. That the mail they're sending is, in fact, spam. 

Email headers don't provide proof of permission. There's no flag in the email message that for sure defines a message as spam (or not). That's why the measure whether or not an email message is spam is so heavily dependent on how the recipient feels about your email message, requiring their input, driven by their perception. Those feedback signals from recipients are invaluable and heavily relied upon.

But I digress. Back to the point: Mailbox providers aren't trying to greenlight spam. They make it pretty clear that what they want to see is permission-based sending. 

Take Gmail, for example. Their published Email Sender Guidelines are very direct on the subject. From Google's own words:

• "If you manage mailing lists or other email subscriptions, you should send email only to people who want to get messages from you."
• "Make sure recipients opt in to get messages from you."
• "Confirm each recipient's email address before subscribing them."

That's not ambiguous. Google requires opt-in.

So if somebody tries to tell you that permission isn't required, you can point them straight to Google's guidelines. Case closed.