Message Bus bets its cloud-native messaging service will improve the art of email delivery

Message Bus bets its cloud-native messaging service will improve the art of email delivery

Summary: Message Bus has a simple goal: help customers keep their legitimate email messages out of recipients' spam folders.

TOPICS: Security

Message Bus has a pedigreed CEO, an impressive list of customers and partners and technology that makes its cloud-based service highly scalable and resilient, yet the young company's goal is simple: help customers keep their legitimate email messages out of recipients' spam folders.

With Twitter co-founder Jeremy LaTrasse at the helm, Message Bus is navigating the often dark waters of email delivery so that its customers don't have to. The company's Global Delivery Network, launched in mid-November, aims to be to email and mobile messaging what Amazon Web Services are to cloud computing and Dropbox to cloud storage.

The service is a cloud-native application, meaning that it's not tied to the underlying infrastructure of a single cloud service provider. Therefore Message Bus can scale and move its customers' workloads across different cloud infrastructures as needed (the company says it currently deploys on Joyent, Amazon Web Services and Rackspace cloud services). This approach avoids the scale limitations of working with a single cloud service provider, as well as the possibility of service disruption if a provider experiences an outage.

But it takes more than the right architecture to provide an effective message delivery service. Message Bus has done extensive relationship building with top ISPs including AOL, Microsoft and Google to understand what they expect from a trusted sender and sticks to those guidelines, resulting in a higher likelihood that legitimate emails make it to the inbox.

"More than 90 percent of all mail worldwide ends up in one of those places; if there’s no trust with those ISPs then the message won’t make it into the box," says LaTrasse. "So we had the idea to build best practices into the network, so everyone who sends through our service follows them. We made the relationships happen, and all our customers benefit, as well as their recipients."

Out of control

Currently, one in five legitimate emails is either blocked or routed to the spam folder, says Message Bus, making it difficult for companies relying on email as a primary driver of revenue and brand recognition to get their message across. What's more, the cost and complexity of launching messaging campaigns across multiple channels (email, mobile and social messaging, etc) is spinning out of control.

Customers of the Global Delivery Network don't need dedicated messaging hardware or personnel; instead they build a virtual SMTP bridge to send their messages across Message Bus' network. This significantly reduces upfront infrastructure costs as well as ongoing staffing, says LaTrasse, and allows customers to focus on the content of the messages, knowing that they'll be delivered in a manner that's effective, secure, and complaint.

At the same it unveiled the Global Delivery Network Message Bus launched a free reporting service called Discover that informs customers of email senders who may be abusing their domain name for illicit or unauthorized purposes. And late in November the company announced an enhancement to its service with the deployment of Opscode's Hosted Chef to automate configuration, environment and application management across the multiple cloud infrastructures powering the company's service.

Message Bus lists American Greetings, MyFitnessPal, and Telly among its early users.

(BriefingsDirect contributor Cara Garretson provided editorial assistance and research on this post. She can be reached on LinkedIn at

You may also be interested in:



Topic: Security

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Great architecture - but where does anti-spam come in?

    Okay, it has a supposedly fantastic "cloud" architecture.

    But where does the anti-spam come in? Who is to say that spammers can't find a way to abuse this? For all of the talk about the spam problem, I don't see a solution here.

    All I see here is "trust the cloud. it's magick. it fixes everything."
    • re: great architecture...

      @cobraa1 hi - i'm responsible for product management & engineering at message bus.

      no matter how sensational the press would be, we do not view our platform as a candidate for FUSSP status. we don't think the cloud is magic, it's just a much better medium for deploying the kind of platform we've designed (at least we think so, there isn't currently a "private cloud" version of our offering).

      like any other service where a client somewhere has credentials to access a protected resource, if that client is compromised the attacker has access to the resource. what we as a messaging platform do that we think is unique is that whereas other services are a permissive, pay-for-play arrangement, our platform is not permissive, it operates by default in a deny-all posture.

      prospective clients must clear a rigorous vetting process before we'll accept them as customers. we will, in the end, be sending their traffic over IP space that we own (not dynamically-assigned churn-and-burn addresses) - protecting IP reputation while guaranteeing message delivery to recipients that want the traffic is the core function of the platform.

      so - what if a client is on their best behavior during vetting and then reverts to poor behavior when they go online, or what if they're compromised and the attacker tries to use them as a spam cannon? a few things stand between the attacker and a successful spam campaign.

      first, every new client is subject to a default sending profile designed specifically to limit their ability to do harm. there are hard limits on the volume they can request us to send and the rate at which we'll send it. well-behaving clients will quickly see sending privileges increase, and to avoid penalizing anyone unfairly we closely monitor their sending practices to ensure they're not unreasonably constrained during this "get to know you" period.

      second, we know what our client's typical sending patterns are, and because we know how much mail we're being asked to send before we send it we can stop it from ever leaving if it falls outside normal, or even peak-normal conditions.

      third, negative receiver-side and recipient telemetry is applied directly to the delivery pipeline in real-time. so if a savvy attacker compromises message content and keeps sending volumes "normal", elevated junk reports and 3rd-party feedback will automatically throttle or pause sending. on the slightest signal that a recipient doesn't want traffic from a sender, we prevent the sender from contacting that recipient through the platform.

      fourth, we are aggressively evaluating outbound spam filtering solutions that bring a battery of industry-leading spam/phish detection to our pre-mta outbound queues. we see no need to reinvent here, leveraging instead what is already proven effective.

      so this really isn't about "solving spam". it's about respecting recipient preferences and operating a high-volume trusted messaging "channel" between out clients and recipient domains such as gmail, hotmail, and all the rest.

      that may not have been a super-sexy answer, but i hope it helps shed a bit more light on how we're trying to change the messaging industry.