I had intended to move on to a topic other than delivery, as I seem to write about it way too much. But a ClickZ commenter posted the following comment to last month’s column that I feel deserves more than a cursory response.
“Gmail’s spam filtering is terrible. in the past couple of months it has become overly aggressive and is erroneously marking legitimate emails as spam. This never used to happen it used to work perfectly now Gmail’s spam filter is terrible. Adding the sender as a contact is an unacceptable solution to the problem. Why would I want to add a an (sic) email address from a non human that is just something such as an order conformation? (sic) That just clutters up your contact list.”
Guest makes several observations and some good points here, but I sense a certain amount of misunderstanding about what Gmail and most major Internet service providers (ISPs) are trying to achieve with their spam filtering.
The first claim is that Gmail is erroneously marking legitimate emails as spam. I must state the caveat that I agree Gmail has been through a somewhat aggressive phase in its filtering. Yet even with that caveat I have to take issue with “erroneously” and “legitimate.” The difficulty with defining legitimacy is that every marketer I’ve ever spoken to believes her email is legitimate. A classic example was seen on “The Daily Show” in 2004. If you’ve never seen it it’s well worth a few minutes of your time. In my own experience I’ve never met anyone who said, “Yeah we’re sending the most egregious crap, no one wants it, but who cares?” I’m sure they must be out there somewhere, but they’re pretty uncommon. Everyone thinks their email is legitimate.
More importantly the claim that Gmail is erroneously filtering legitimate email suggests to me that Guest believes it is Gmail’s aim only to filter illegitimate email, but this is a misconception. ISPs long ago realized what I stated above – that almost every marketer believes their email is legitimate. So they gave up trying to separate the legitimate from the illegitimate, the opt-in from the opt-out, the Viagra from the Viagel. Instead they accepted the maxim that spam is “any email I don’t want” and changed their goal to filtering based on what their recipients want.
My advice to Guest therefore is, in addition to looking at the filtering reasons given by Gmail, change how you judge Gmail’s filter. Rather than deciding how good or terrible it is based on how you view your email, consider how your recipients view your email because that’s what Gmail, Yahoo, Hotmail, AOL, etc. are doing.
You can do this by measuring how recipients are engaging with your messaging. Negative indicators are bounce rate, complaint rate, and unsubscription rate. Positives are open and click-through rate, and possibly also rates of social sharing and forwarding. Regardless of how legitimate your messages are, if your positive indicators are low and your negatives high, Gmail will be filtering your email, and that is exactly its intention.
I’d also like to comment on the “add to address book” solution. I agree completely with Guest about adding non-human senders to address books. I know that recommending users do it is a best practice, and I know that it helps to ensure your email isn’t filtered, but I have yet to see any good data on how widely it’s done and the solution simply doesn’t seem capable of operating effectively at scale.
The bottom line is that if you want to be successful at getting to the inbox, start by understanding the objectives of the gatekeepers and the rules under which you’re operating. Even the best senders get filtered sometimes and on occasion even the best filters act terribly, but the major ISPs get it “right” far more often than most marketers would like to believe, and blaming the ISPs is not an effective delivery strategy.
Until next time,