It’s just possible that they are acting with the best of intentions, but I for one have always had major concerns about the way email providers – Google, Microsoft et al – deal with the issue of spam. In my view, however well-intentioned, interfering with someone’s email infringes their rights as citizens and is potentially illegal. And to my mind, things are about to get a whole lot worse.
Google’s imminent launch of its “Gmail Priority Inbox”, which analyses your behaviour and “decides” which emails are NOT important, and the news that Microsoft is in the early stages of utilizing user behaviour to decide whether a message is delivered to the inbox or junked for Windows Live Hotmail, bring the likelihood of a class action lawsuit against them that much closer.
As I understand it, in the US it is a Federal crime to hide, keep or open others people’s (snail) mail without permission. The same is true in the UK, of course, where rather quaintly all mail in transit is considered to be the property of the Queen until it reaches the owner's mailbox, at which time it becomes theirs. In both countries, interfering with the mail is a serious offence and could result in a jail term and a large fine. The law does not care whether the letter is a utility bill, bank statement or yet another unwanted piece of direct mail.
If something is sent to you in a sealed envelope with your name and address on, that communication is considered sacrosanct; it is yours alone to decide if, when – or whether to read it. With email, things are rather different, even though an email address is every bit as central to people’s lives as a home address.
Today an email address is at the heart of a person’s identity. Without it, most of our online transactions and interactions – with individuals, brands and shops, with government and employers – would not be possible. In the developed world, not having an email address is the online equivalent of being homeless.
In other words, to have an email address and unrestricted access to your email has become a basic human right. Rights are things that confer responsibility on others. In this case, email can no longer be regarded as the property of the providers. The decision as to whether I receive an email from a company operating within the law, whose newsletter I have subscribed to should not be in the gift of the people who happen to run the delivery mechanism.
In summary (for those of you who have been on another planet the last couple of weeks) the headline news is that Gmail are going to start prioritising the emails they deliver to you based on engagement – the way you interact with emails. They will be looking at variables such as:
1. How often you read the email from a given sender
2. How often you reply to the email
3. Manual adjustments using the increase or decrease importance buttons
Hotmail are going one step further and are planning to junk email that doesn’t meet their as yet to be defined engagement standards. Metrics Hotmail will be taking into account include:
- Messages read, then deleted
- Messages deleted without being read
- Messages replied to
- Frequency of receiving and reading a message from a source
On the basis of such metrics, Google and Microsoft will start deciding that emails that you’ve signed up for and are sent legitimately should be deemed junk or unimportant, because the way you’ve reacted to them doesn’t fit with their idea of “engagement”. Now no system is infallible and systems that rely on interpretations of the meaning of a given behaviour are particularly vulnerable.
An obvious problem with this is that, despite what the email experts say, indifference to marketing and advertising activity in in all channels is actually the norm and email is no different. If someone sees an ad from a holiday company when they are not planning or thinking about a vacation, their most likely response is to ignore the ad, turn the page or leave the room to get a cup of tea and certainly not to book a holiday. This does not mean that they hate the company and never want to hear from them again: in fact the paradox is that email marketers – who operate in the only channel where you can actually opt not to get messages from companies you don’t like spend more time worrying about being relevant than marketers selling feminine hygiene products on TV!
So the danger is that the new Google/Microsoft initiatives will effectively classify the majority if not all commercial messages as second-class post. It’s the equivalent of the postman (or woman) delivering some of your mail to your mailbox, but leaving your utility bills in a cardboard box on your doorstep, out in the rain, because they heard you complain about your phone company yesterday.
You’d think the email marketing community’s response would be outrage and anger. But no, it’s: GREAT the bar’s been raised even higher! We’re going to have to be even more engaging and relevant than ever! Batch and blast is truly dead, hooray! Talk about turkeys voting for Christmas.
It’s a classic example of fear and self-loathing in email marketing! Our community, “guided” by people with a vested interest in, and/or a career built on, the idea of inbox delivery being difficult, are implying that these new developments serve us right because we’re not being relevant. ESPs, deliverability companies, and the vast majority of email agencies out there operate by this mantra: if something doesn’t reach someone’s inbox, it’s our fault. No one is questioning whether the ISPs’ latest response to the spam issue is either reasonable or legitimate.
Anyone who thinks that the either Microsoft or Google are on the verge of eliminating spam for ever is just plain nuts. Whatever they do, Viagra, porn, Rolex watch and phishing emails – false negatives will still occasionally appear in our inbox.
What should be of concern is the fact that inbox delivery based on behaviour will significantly increase the false positives – email from friends and co-workers, order confirmations and delivery notifications that get marked as spam or deemed unimportant. And this likely to get worse as more and more people get smartphones
A lot of people perform “email triage” on their smartphone: they use their mobile inbox to filter, prioritise, read and/or delete messages. And the phone may or may not enable images. So if you do email triage on your Blackberry (which defaults to no images) and delete messages after you’ve engaged with them to your satisfaction, Gmail and Hotmail may well never know. They may just see that as damning evidence of non-engagement and start blocking those messages. Mobiles will make it very difficult for Google and Hotmail to truly understand how people are engaging with your email.
So there are still problems in need of a solution. But rather than challenge the right of ISPs to vet people’s email, we beat ourselves up and buy in expensive deliverability expertise to help reach the inbox of someone who has asked you to send them email.
No one is pointing to the increased risk of making the wrong inferences from a given behaviour – of assuming that“I don’t open an email” = “I never want your emails”. Many people rarely open their bank statement; imagine if, after 7 consecutive instances of me not opening my statements, the postal service decided to stop delivering them! People who do online banking may go several months without opening emails from their bank, and they can do that safe in the knowledge that those emails will never be junked or deleted from their inbox; they’re there forever. If they got junked or blocked by Hotmail or – under the new Google system – “deprioritised”, there is a danger that someone will not see or delete an important message in the belief that it’s something spammy they don’t need.
One metric of engagement apparently being used by Hotmail is replying to an email. But very few people reply to commercial messages or bank statements or order confirmations (in fact the from address is often “donotreply”). It’s not the only criterion but it will count against you as an email marketer. Imagine the outcry if this happened to our offline mail service.
Given that we’re quickly moving to a place where email will be considered to be a valid and legally binding communication channel in its own right, what’s going to happen soon is that an individual is going to miss something vital. One day – if it hasn’t happened already – emails missed because they are falsely branded as spam will cause someone to miss an appointment for an operation, lose a contract or miss a mortgage payment.
Legal advice suggests that there might well be a case for someone to sue an ISP for failing to deliver a message. For obvious reasons, this would need to be a class action, and one can imagine several classes of people who might dare take on the ISPs:
1. Individual claimants aggregated by opportunistic lawyers who’ve suffered from non-delivery with the prospect of a huge jackpot
2. Small businesses that use Google’s hosted-email service, which operates by the same rules
3. Very large brands: legitimate businesses operating within the law; the very same businesses who collectively spend hundreds of millions of their advertising and marketing budgets with those very same ISPs (and even if they chose not litigate, they could – should – threaten to withdraw their advertising)
4. Social networking companies, which rely heavily on email to keep their less than totally addicted users engaged with what’s happening to their friends and drive people back to the site (If you were Facebook or Yahoo groups would you trust Google’s behavioural algorithms to be impartial?)
Because the behavioural approach is such a black box – there are no clear definitions of engagement – it gives big ISPs the opportunity to be anti-competitive – to favour some companies over others. And so even if they are going to use this kind of algorithm they should be obliged to publish exactly what the algorithm is.
I think we should stop using deliverability as a yardstick for the effectiveness and quality of our email marketing activity. It is an artificial and constantly moving benchmark developed for the convenience of the ISP’s. It very rarely – in the case of permission based or opt-in programs run by trusted brands bears any relation to conversions and revenue
Where we are with spam blocking though imperfect goes far enough, the problem with the Google/Microsoft approach is that they are moving beyond the wholly laudable project of trying to stop illegal activity, and to a totalitarian state where they decides who reads what.