You know, I'm pretty sure I know spam when I see it, and it would seem most people did as well. Defining exactly what it is, though-- that's a harder task (didn't Justice Stewart once say something like this about another dark corner of the internet?) Congress did pass CAN-SPAM in 2003, which provided regulatory guidelines as to what legally constituted spam, but what the law says isn't necessarily how consumers see things. From their perspective, spam can often mean any unwanted email-- whether it's commercial or not.
Unhappy when they receive spam, consumers put pressure on ISPs and email providers, who then block email marketing based on stricter guidelines than dictated by CAN-SPAM. As the FTC's recent report to Congress showed, the number of blocked emails only grows every year, and AOL alone blocked half a trillion messages last year. This number most definitely includes many missives that met legal and industry requirements for opt-in, disclosure and opt-out procedures.
This may seem a bit like overkill. But ISPs have a legitimate interest in keeping their subscribers happy, and without clear cut lines between what email is desired and what is not (remember, we're using the consumer definition of 'spam' here), they're better off doing more rather than less. Then again, this can be expensive, and consumers may also complain when they don't receive a message they wanted. In the last year, ISPs and email marketers have started turning to two different solutions to navigate this environment and differentiate spam from legitimate email marketing:
- Authentication and Reputation. Using a "Sender ID" or "DKIM" protocol, ISPs work to determine whether or not an email comes from the internet domain it claims to. Beyond this simple identification, some companies also rank the reputation of email senders or outgoing domains, dependent upon email practices and complaints of spam. Some ISPs use this as a rough guide to legitimacy, and place certain senders on a whitelist, allowing images and links through their spam filters.
- Third party certification. Similarly to how websites are verified as secure or considerate of consumer privacy, some companies have begun accrediting email senders as legitimate. These companies work with senders to make sure they are legally compliant and meet certain standard practices. In return, ISPs agree to let email from these 'trusted senders' load images and links-- and go straight to consumer inboxes.
For a while, it seemed that the first method might gain acceptance. While the use of reputation standards has lagged in the past, the Email Senders & Providers Coalition reports that 70 percent of Fortune 100 companies now employ at least one such procedure. ISPs like AOL have developed whitelists of legitimate senders and firms such as ReturnPath work with clients to improve their email practices, keep spam complaints down and boost delivery rates. At the end of January, though, AOL announced that it was phasing out its "enhanced" whitelist and moving towards using a third party certification provided by Goodmail.
AOL argues that certification will provide a "safer and more secure environment for our consumers, and restore some trust in the email inbox." For its part, Goodmail believes that certification will improve the success of legitimate marketing while continuing the focus on consumer permissions. Although specific reputation requirements haven't been published, Richard Gingras, Goodmail's CEO, told me that the CertifiedEmail prerequisites are strict enough that even "many exceedingly brand-conscious Fortune 500 marketers will not qualify for participation in the CertifiedEmail program due to unacceptable reputation scores."
Not everyone agrees with AOL's move, however. Goodmail charges its certified senders by the email sent-- essentially charging email postage (as they explicitly suggested back in 2004). Matt Blumberg, CEO of Return Path, commented that Goodmail's CertifiedEmail program, by charging even a "fraction of a cent" per email, could double the cost of commercial email. This means, according to Blumberg, that although "some mailers like high-value marketers will be able to afford that cost, many others, such as content publishers and non-profits, will not be able to afford the price at all."
Even so, it is possible that some marketers might benefit from the use of third party certification. As ReturnPath themselves reported in January, consumers are increasingly hitting the "this is spam" button instead of unsubscribing from opt-in marketing. According to the survey, 34 percent of email marketing recipients said they report messages they don't want as spam to their ISP. In an environment like this, it's arguable that many marketers would prefer certification over a whitelist based on reputation, where even a few "this is spam" clicks can mean removal.
In addition, there might be greater reason to track consumer permissions: Jupiter Research's David Daniels has suggested that if marketers were to pay per email sent, they would have an incentive to closely monitor unresponsive email addresses-- thereby limiting messages sent to consumers who don't want them.
Remembering the consumer
That being said, there remain important questions about where the distinction between spam and legitimate marketing should come from. At least part of the value of reputation-based whitelists is that they give a level of control to the consumers whose inboxes we're talking about. The point here is that we want to make sure that good reputations are rewarded-- the "this is spam" button should have some effect. Third party certification could be part of this-- and help to take the weight off of ISPs.
By charging email marketers, AOL does stand to recoup a significant portion of the money they've spent developing spam filters and other security measures. And as long as the third party certification retains an emphasis on consumer permission, then AOL will continue to have strong reason to keep its primary customers -- subscribers -- happy. The concept of charging for email delivery in and of itself may even make sense-- even if it does push some of the more fringe companies out.
One of the best things about the internet is the entrepreneurialism it encourages. The flip side of that statement, though, is that there is a significant minority in this space that has little skin in the game. And perhaps assigning some additional cost to email would force all of us to more thoroughly evaluate the emails that we're deploying. The key question, of course, remains "how much additional cost."
With that in mind, I'm somewhat uncomfortable with the lack of transparency around the CertifiedEmail program. For example, it isn't clear how much senders will be charged per email -- we've heard "a fraction of a cent," and the New York Times has estimated the cost at somewhere between ¼ cent to 1 cent per email -- and who is in charge of determining that cost. Similarly, as long as Goodmail's reputation requirements remain unpublished, it's rather difficult to evaluate just how effective those requirements are.
In a strange twist, after some initial controversy, AOL has now announced that it will retain its "enhanced whitelist," at least through 2006 -- while still phasing in the CertifiedEmail program. With both options available to them, many marketers may want to wait and see how the certification plan works out-- both in terms of cost and reputation standards. Ultimately, respecting consumer permissions is still the best strategy for making sure email marketing arrives -- with functional links and images -- in consumers' inboxes, whether this is done through whitelist inclusion or third party certification.
Isaac Scarborough is manager of market intelligence at Chapell & Associates. Read full bio.
iMedia: On the creative front, widgets appear to be the natural evolution of rich media display advertising. There are already established players in this space. How are you integrating with those platforms?
Radfar: Widgets served as rich media is clearly where things are going. We are a partner-driven company and believe that by working with leaders in the advertising and reach media space, we can make our clients more successful. As such, Clearspring has established an exclusive partnership to integrate with PointRoll to create our SnaggableAd product. With PointRoll owning nearly 70 percent market share in rich media online advertising, we think that we can create some serious value. In addition, we have also integrated with most of the leading ad networks and ad servers including DoubleClick and MediaPlex.
Pashman: Clearly rich media ad units are a natural starting point for branded widgets. We announced the first deal in the marketplace back in November with Eyewonder. Our Wildfire technology enables users to install content directly from an ad unit -- basically converting an Eyewonder ad unit into a widget. We expect to see lots more activity in the rich media space throughout the first quarter of this year.
iMedia: We're starting to see more examples of quantifying the momentum effect enabled via widget-based campaigns on social networks. What are your firm's capabilities regarding tracking interaction/engagement?
Radfar: Since creating the first viral analytics hubs for widgets in 2006, we have invested heavily in understanding widget distribution and have developed patent-pending technology in that area. We understand what the affinity distribution paths are and then visualize a topography that identifies the hubs responsible for the higher performing through-put.
Metrics standards are critical for the success of any space. As such, we are working tightly with folks in the online measurement space to standardize viral metrics like these, as adoption of metrics is crucial to the success of the ecosystem.
Pashman: Gigya is tracking approximately 2 billon widget impressions per month. We track stats like total posts and impressions -- both of which can be queried by day -- and date range as well as by unique ID. We also track viral distribution patterns so advertisers gain insight into where their ad is getting traction. Because widgets allow advertisers to interact with their audience in new ways, we also give advertisers the ability to track any interaction metrics they want. So if the widget has three videos and six MP3s, the advertiser can give us nine different tags and track the interaction of the widget and what songs/videos are most popular. There is no restriction on the amount or type of interaction metrics an advertiser can track inside the widget.
iMedia: Are marketers "getting it" for these types of analytics?
Radfar: I think so. Clearly CPM is not going to disappear right away, but there is definitely a trend toward embracing new models and metrics. We are engaged with web metrics experts at all levels. Those parties, as well as our customers, are definitely asking the right questions -- they want to understand how you could look at particular data on a page. They're interested in the fact that we're calculating time spent, grabbing all of the clicks and how to instrument their widgets to collect custom events, like unique sequences of interaction.
Keep in mind that for the advertiser, there are analytics about the widget, where it's going, what it is doing, etc. And then you have an ad tag, potentially, that's served into the middle of that widget that's capturing different metrics. And that's where I see our value, pulling in that unique widget engagement data versus the data that the ad tag will produce. In the near term, I think we're most likely going to see variations on CPC/CPA type models applied to widget-based ad buys -- a CPI (interaction) if you will -- because whatever is simplest will be the most used and marketers are comfortable with pay-for-performance metrics. But realize that the way in which companies measure, value and pay for widget campaigns is changing rapidly.
Pashman: I think the question you are asking is, "How are advertisers defining success" for these kinds of campaigns? Even in traditional campaigns, hitting CPC or impression metrics doesn't mean a campaign is successful, but experience shows that hitting certain metrics is likely to impact brand awareness, sales or other business and marketing goals. We think it's important to work closely with advertisers to define campaign success in the early stages, both to figure out how these campaigns can achieve some traditional brand goals, but more importantly how to achieve other goals -- new brand insights, identifying and getting feedback from real brand advocates.
Larry Everling is the president of Grady Rose Consulting. Read full bio.