What if ESPs (and ISPs, for that matter) started charging users for every complaint generated? Think of it like peak pricing for electricity. In California, businesses can opt for discounted power, with the agreement that they are the first companies shut off if electrical demand exceeds supply. What if ESPs and ISPs offered discounted hosting rates to bulk senders who agreed to pay per complaint?
I see pricing scheme something like this.
- $5.00 per FBL message.
- $50 for a hand written complaint.
- $150 for a report of a spamtrap hit.
- $500 for an ISP temporary block.
- $10K for a major blacklisting (SBL, other filtering company).
- $5K per customer if the blacklisting affects other customers’ ability to send.
ESPs could give a threshold of complaints that are covered. For instance, every per-complaint customer gets 0.05% of their total volume in free FBL complaints. Hand written complaints they get one or two every billing cycle, not to exceed 12 complaints a year. 6 spam trap hits a year. There is a bit of grace in the handling. I can think of lots of ways to make this sender friendly.
Benefits to the Sender
Good senders benefit because they get lower rates and don’t risk much in the way of complaint related expenses. They don’t have to pay that hidden compliance fee that all ESPs customers currently pay.
Senders with more aggressive email programs benefit because they’re able to shoulder the risks associated with those programs directly. The ESP has less to say about buying or renting lists because instead of the ESP paying the cost of problems, the customer assumes that cost. Even better, the sender can pass the compliance cost back onto their list vendor. Imagine being able to tell a vendor they have to discount a list based on the number of complaints or delivery problems. Instead of the vendor selling a list with no incentive for that list to perform well, the vendor now has an incentive to make sure those opt-in addresses really are opt-in.
Benefits to the ESP
The ESP benefits because no longer is their abuse and compliance desk a cost center. With the right mix of clients it may even become a profit center. Plus, the compliance desk is guaranteed to be funded at a rate that covers the work needed to maintain a reasonable delivery rate for customers. The ESP can stop spending so much time telling customers they can’t do something and more time raking in the cash.
It’s a total win for everyone.
Monetize complaints. It’s the future of email.
RSS - Posts
This is a very interesting perspective and one that deserves merit. However, competitive dynamics between ESPs make this idea unfeasible. Given that there are over 100 ESPs it would require a commitment by almost all of them to abide by this proposed or similar fee structure. Otherwise, all it takes is a few ESPs to break away from the idea for competitive reasons and the idea completely unravels.
I see a few of issues with this.
1) This would require revealing the full contents of complaints to the sender, which is something that’s generally discouraged. I realize many ISPs/ESPs do this anyway, but you realistically cannot bill someone for something they’re not allowed to see.
2) This might provide an incentive for a company to not deal with minor abuse issues at all, preferring $10K for a blacklisting over a bunch of one-off complaints.
3) Collecting from a spammer will, as always, be problematic.
4) FBLs are extremely noisy if you don’t have a significant enough volume to generate a usable stream. I have users who generate more than 0.05% of their total volume in FBL complaints who send no bulk mail at all; the FBL messages they get are virtually always cases where they replied to a regular correspondent and that correspondent smacked the “This is spam!” button on a reply that they requested.
#4 can be fixed by fine tuning the numbers, fixed minimums, or by allowing disputes, but the whole thing turns into a mess.
Still, it’s not a terrible idea if an ESP can make the numbers work, and can survive the onslaught of “This ESP will let you pay to spam!” (which itself isn’t entirely incorrect)
If ISPs were amenable to this type of monetization, then Goodmail would still be around…
This is a good start, but you’re going to need a more sophisticated pricing formula that deals with the cost of IPs you burn and so forth.
We’ve been hearing for years about dynamic power pricing, in which customers and generators bid in real time for increments of power. I’d think the same thing would be applicable to spam.
(cough) Bonded Sender (cough)
I don’t really think my idea is the same as Bonded Sender or Goodmail.
This is the ESPs charging directly for the amount of work a particular customer creates for them as a service provider. Bonded Sender and Goodmail were 3rd parties charging for access to receiver ISPs.
Companies that agree to the lower price point can keep their costs low by managing their email programs in a way to minimize complaints. Companies that want to send questionable mail can pay more for the privilege.
It’s really about the ESPs finding a way that their compliance desks don’t have to be “the big, bad people who say no to senders who just want to market effectively.” It’s also about making the customers absorb the actual cost of their marketing decisions. You want to buy a list? Sure, but you’ll have to pay for every complaint that list generates. If that list causes major problems (a block), then you’ll have to pay even more. If the list causes a widespread block and affects other customers, then those customers are compensated for their delivery problems by the company that directly caused them.
I expect, actually, that once senders have to bear the direct costs of their bad behavior they’ll be much less likely to do stuff outside the realm of best practices. Right now, there is really no cost to buying a list outside the list itself. The cost of the bad send is paid by the ESP.
Hey Laura,
I like it… We even have been in talks internally to implement a plan to monetize from the spam complaints generated by our users.
However, there are some issues I see:
1. Hard to collect from senders (especially amounts over $500)
2. Most ESP’s have to be on board or else senders can say “I can go to another service and they’re cool with it…”
Great job though 🙂
I think what people are missing here is that it’s just a different pricing model. Adobe switching to record-based from volume-based pricing didn’t either put everyone else out of business or require everyone else to change. Neither does this.
Whether this would work or not depends on the details of the value exchange. If the ESP is able to reduce their costs or make their revenue more directly proportionate to their costs it may be a win. I may be a double win if it causes unprofitable/expensive clients to go elsewhere.
On the other hand if it just enables clients to pick their own lowest-cost option and they do so effectively then it will likely be a net loss for the ESPs.
This also assumes every Spam complaint is valid. We all know that a decent percentage of them are generated by recipients who either:
1) Forgot that they originally subscribed to something; or,
2) Don’t follow the proper opt-out process
So, to charge senders without undertaking some additional due diligence as to a complaint’s reason doesn’t seem fair…
Russell I think that is why you give them a % that are allowed to complain as the ISPs do. As Laura wrote above you give them a .5% leniency for this margin of error. We already do a hybrid of this model internally.
Also, the larger point here is that, in order to know which spam complaints are valid, someone has to investigate, that investigation has a cost, and that cost is currently passed on to senders. So, they’re ALREADY paying for it. Shifting the model toward putting those costs up front allows the senders with low complaint volumes to pay less, and those with high complaint volumes to pay more. For the senders, this is more transparent, and (if they’re doing the right things) cheaper, and for the internet at large, it’s an incentive for senders to do the right things. Everybody wins, except the bad guys.