There’s a battle being waged on the interactive targeting front. No, it’s not the adware vs. spyware dilemma; that’s a matter for Capitol Hill. The online siege I am referring to is the ongoing and evolving battle advertisers face when trying to establish parameters for beyond-the-box targeting.
In this mini-war there are two opposing forces, standing on opposite sides of the audience-targeting battlefield: the behavioral targeting freedom fighters and the contextual armada. Both present valid and strong arguments for reaching audiences with sought-after messaging within critical desired action time frames. Unfortunately, comparative results data arguing for either side remains largely inconclusive.
The real debate seems to be the implementation tools in targeting methods. Privacy and the disclosure of Personally Identifiable Information (PII) and the ongoing debate about the future of cookies (are they or are they not going away?) are at the forefront of this war on targeting. Let’s see if we can intervene and try to find some common ground, formulate some documentation for a peace treaty and help make the online world a safe place for both parties.
Behavioral says what?
Recently, this topic came up in a heated debate between yours truly and Bill Gossman, chief executive officer of Revenue Science, in an industry guru panel for financial analysts. I later caught up with Bill and Omar Tawakol -- Revenue Science’s senior vice president of marketing -- to continue our discussion.
Since an audience poll I conducted at a recent conference and subsequent discussions with conference attendees indicated that few people could provide a cogent explanation of behavioral targeting, I thought we’d start there.
“The first and easiest definition of behavioral is targeting people, not pages or actions,” Gossman says. “The best way to target consumers is to be where they are and not interrupt their experience by mapping the content to the user.”
Revenue Science has carved out a place for itself in the online world by doing just that, teaching publishers, content owners and advertisers a smarter way to reach out to its user base by using techniques that include maximizing user registration data and, of course, cookies.
At the heart of Revenue Science’s technology is the proprietary Audience Search interface that, ironically enough, uses keywords (along with rules-based criteria that includes behavior and/or registration data) to help identify a potential audience. “The person provides the context and the place,” reports Tawakol. “If done correctly, behavioral targeting is not intrusive, respects privacy, and does not interrupt the user experience.”
OK, we get it: Reaching out to users based on what you know about them appears to be a hell of a lot better than simply pouring a bunch of advertisements (text, or graphic) out on the proverbial stoop to see if the neighborhood cats lick them up.
Shelving the privacy and cookie discussion for a moment, the debate seems pretty well defined now. Is it better to serve an ad on the basis of what you know about the user or the content on the page?
Contextual come back
Contextual targeting with graphical- or text-based ad units relies on using search terms to deliver ad units to users based on self designated criteria -- in and of itself a tremendous burden for contextual targeting providers. A defined need for publishers to generate revenue with advertising got a stiff kick in the butt when search providers started to integrate text ads into pages based on directive queries.
On the heels of directive search sponsored listings success, contextual networks -- as they came to be called (e.g. Google’s AdSense) -- allowed publishers to place sponsored listings auction style on content pages and share in click-response-based revenue. Putting it mildly, advertisers quickly found that response behavior wasn’t quite as good in contextual listings, and the advertising world cried out for something better.
The content construct
The foundation of most contextual search listing taxonomies is keyword scan technology that matches text or graphic ads with page content. Even that didn’t seem to be good enough, so other methods of implementation were constructed -- such as sophisticated mapping technologies that did not rely solely on keyword scans but created more effective categories for advertisers to select ad placements, Chinese menu style.
Quigo Technologies is one such innovator in the arena of content-based advertising. Quigo’s AdSonar platform offers a wide array of keyword-based content targeting options for publishers and advertisers. “The two most interesting pieces [of AdSonar] are topic based -- to the tune of about 5,000 topics," says Yaron Galai, Quigo’s senior vice president of product development. “The system decides what response-based vehicles are most appropriate, the most interesting is the private label component.”
Offering private labeling contextual search -- such as the offering here with cars.com -- and then balancing the publisher need to protect rates with the advertiser’s desire to select individual sites in which to place contextual advertisements adds up to an area in which Quigo excels. Advertisers can elect to place listings on specific sites, a unique and much needed feature in contextual search.
More questions, not enough answers
Clearly, the winners of this war will be those who can go beyond the directive search realm and into the mind of the browser via the vast uncharted content abyss. The big question still remains unanswered. Is a behavioral platform better than a content and keyword driven one? To answer that query, we’ll have to spend a little time with privacy issues and take a closer look at response behavior. You’ll just have to wait until next week for the rest of the story. Stay tuned: The answer just might surprise you.
iMedia Search Editor Kevin Ryan’s current and former client roster reads like a “who’s who” in big brands; Rolex Watch, USA, State Farm Insurance, Farmers Insurance, Minolta Corporation, Samsung Electronics America, Toyota Motor Sales, USA, Panasonic Services, and the Hilton Hotels brands, to name a few. Ryan believes in sound guidance, creative thought, accountable actions and collaborative execution as applied to search, or any form of marketing. His principled approach and staunch commitment to the industry have made him one of the most sought after personalities in online marketing. Ryan volunteers his time with the Interactive Advertising Bureau, Search Engine Marketing Professional Organization, and several regional non-profit organizations.
Meet Kevin Ryan this week at Ad:Tech San Francisco.
Return to Award-winning emails
A fundraising email from WBFO, a Boston based public radio station.
This email was sent about five years ago, but I have kept a copy because it's a great example of two separate points.
First, and this is a point we can cover in just a few lines, this is an excellent example of what I call "HTML Lite." That is to say, the email has all the advantages of using HTML, but also uses the personal nature of text-only to carry its message. The use of HTML allows them to use their logo in the header area and then place hyperlinked text in the body of the email, instead of using long, ungainly text URLs.
By resisting the temptation to use HTML to over-format the page and add numerous irrelevant graphics and links, they have retained the personal touch you get with a text-only email
It is rare to see HTML Lite. And that's a pity, as it gives you all the benefits of HTML and text-only within one email.
On to the next point.
The opening paragraph, just three sentences long, is one of the best I have ever seen.
Here it is:
Listening to great programming on WBFO while sipping a big mug of coffee is a daily ritual for many people. How about you? I'm writing to invite you to be part of the WBFO 200 pound coffee challenge.
We'll take this sentence by sentence.
The first sentence connects the radio station with the topic of the email: coffee. There is no fluff, no clichés about how much we value you as a reader, no warm-up intro. It gets straight to the point. Immediately. And that's rare in an email. But the sentence does more than simply set up what follows. It also paints a picture...a picture of people sitting together at breakfast, enjoying their coffee and listening to public radio.
It's a nostalgic, attractive and comforting picture. It has a touch of Norman Rockwell to it. Not bad for a first sentence.
Now for the second sentence: "How about you?"
Brilliant. The writers (Betsy Harman and Joan Wilson) immediately invite the reader into the cozy picture created by the first sentence. They are including the reader. They are inviting the reader to participate...to imagine, to feel, to enjoy the warmth of the coffee and the sound of the radio.
And they do it in just three words.
In addition, from a technical point of view, it's just perfect that they use a very short sentence in-between the two longer sentences before and after.
Now for the final sentence: I'm writing to invite you to be part of the WBFO 200 pound coffee challenge.
Thank you Betsy and Joan. As a reader, it is an enormous relief when the writer gets straight to the point and explains the purpose of the email right upfront. Most writers, through sneakiness or just carelessness, often try to lead you down a tiresome, winding trail before actually getting to the point.
As for the rest of the email...it works just fine. And even though I don't think the rest of the email matches the mastery of that first paragraph, it doesn't matter.
They had me at "How about you?"
I love the emails I receive from Innocentdrinks.co.uk. I have never tried their drinks, but there is a refreshing innocence about their emails. I like how they choose their topics. I like their use of images. And I love how they write the copy.
Consider the first two lines: "Hello there. This week, by popular demand, we bring you more Shilpee, more bacon, more pinboard and the magic van. Exciting."
I have no idea what they are talking about. But there is an energy and mystery there that makes me want to find out.
Here's the opening from another one of their emails: "Hello. The sun is out, the sky is blue, there's not a cloud to spoil the view, but it's raining, raining in my heart. All true, apart from the raining bit. We're actually feeling quite chipper today."
You begin to ask, "Who ARE these guys?" And you want to find out.
Here's what I like about the copy. It feels like it was written by someone who is totally and genuinely enthusiastic. More important still, it feels like the writer, mercifully, has never taken a copywriting course or read a book on writing to sell.
And what a refreshing relief that is.
You don't need to be on your guard as you read this email. You don't need to activate your BS filters, even unconsciously. It's like listening to a child bubbling over with excitement about some new discovery in life.
To put it another way, it's like sitting down for coffee with someone who is a huge fan of these drinks and happens to have some snapshots in her purse. She chats merrily away about the drinks and presents the snapshots one by one. She's not trying to make a sale. She just loves the drinks and the company that makes them.
This level of "innocence" is very rare in email marketing.
Usually emails are professionally crafted, carefully written to engage the reader and generate a clickthrough. But as soon as one reads an email that is crafted consciously to generate a clickthrough, a small cloud is cast over the communication. Our intent as marketers is unveiled. It is understood that the writer wants us at least to click something, if not actually make a purchase. We apply filters. We are cautious. And the more formal, corporate and deliberate the email is, the more cautious we become.
The thing about the innocent drinks emails is that they do come from a company. And the company does want us to buy their drinks.
But someone in their marketing group was smart enough to say, "OK, we'll do an email campaign, but we won't burden our emails, or our readers, with a sales message. We'll just present ourselves as we are. We'll go for a long-term relationship, not a short-term sale."
I'll always read their emails, because they are always entertaining and never burden me with having to make the decision whether or not to make an immediate purchase.
A follow-up email from Rentalo.com. This may strike you as a strange choice for a "great" email. The personalization is incomplete, there is a carriage return missing between the first two paragraphs, and it's not particularly well written.
So what do I think is great about it?
First, the timing.
During the course of some work I was doing, I signed up at the site as a property owner. The site lists properties around the world that are available to travelers for short-term rentals.
So if you are traveling to Paris, France, but don't want to stay in a hotel, you can go to Rentalo.com and find an apartment.
Anyway, I received this email about a week after I signed up at the site. Immediately after signing up I got a welcome email. Then there was a break of about a week before I received this one.
It was smart of them to wait for a week. I didn't feel they were pushing me or crowding my inbox.
All too often companies flood you with communications as soon as you give them your email address. They can't wait to "leverage" the new relationship they have with you.
Well, that relationship is paper thin and easy to damage.
Rentalo.com made a smart move by giving me a week's breathing space between emails.
Second, the message.
There is only one important thing they needed to say to me a week after I signed up as a property owner. They let me know that people were looking to rent properties in my area. This may sound like a no-brainer, but what they did was very smart. They identified the ONE thing that I, as a reader, WANTED to hear.
They didn't need to write a long sales message. They didn't need to make a song and dance about it. They just found the one thing that would attract my immediate attention and interest.
Take a look at your own emails and ask yourself: Have we identified the ONE thing that this reader WANTS to hear?
Third, the absence of unnecessary text.
Having identified the "one thing," it was probably tempting to add some other points to the email. They could have told me how great their company was doing. They could have tried to upsell or cross-sell me. But they didn't. They simply mentioned what I wanted to hear and briefly told me how to take advantage of the opportunity.
Nothing more. Nothing less.
By getting the fundamental message right, it was easy to forgive them their sloppy formatting, personalization and writing. The execution didn't really matter because the message was dead on.
Isn't that the core of what good marketing is all about? Great marketing offers you something you really want. Mediocre marketing tries to persuade you that you want something.
As I mentioned in the opening to this article, it isn't easy to find examples of great emails. Fortunately, I needed to find only three examples. If I had needed to find 10, I'm not sure I would have found that many.
The problem with most promotional emails is that they are too heavy-handed.
One's email inbox is a personal space. It offers marketers the opportunity to connect with subscribers and customers on a personal, one-to-one basis. In their different ways, each of the three emails above used a light touch to reach their readers, whereas most promotional emails charge into people's inboxes with the subtlety of a late-night infomercial.
That's the wrong approach and misses the opportunity to develop long-term relationships with your readers.
Nick Usborne is a leading authority on the subject of writing for the web. Read full bio.
Why this matters
- Without complete control over placement, networks cannot guarantee that they are not exposing their clients to undesirable/low performing "reach" (think below-the-fold or deeply chained "gallery" exposures) -- where is the value in that?
- How can a vendor guarantee priority of delivery when it is acquiring inventory through a third-party rather than directly from the source? How can that vendor guarantee that it is getting "the best" inventory from the underlying publisher?
- Would any advertiser actually want to reach 100 percent of everyone online and pay for it? What matters most is how consumers are targeted and how and where your ads connect with them.
Ask, "How do my vendors acquire their audience reach?"
- Do they procure the leftovers being handed out by other networks/brokers?
- Do they have the relationship and degree of control required to identify and secure the most desirable inventory on a site?
- Do they procure inventory on a lowest-price basis from exchanges?
- From what sites is the vendor's reach derived? Rich, immersive consumer and brand friendly environments? Social media? Can the source be guaranteed? Can it be vetted?
- Are the impressions that make up their reach sourced at the beginning of users' browsing sessions or near the end (long tail impressions)?
If you aren't comfortable with the answers you receive, move on. There are 400+ other options available to you.
Lesson 2: Potential vs. actual reach
To date, the reach figures quoted by third-party data providers represent a network's "potential" audience. And potential audience reach is different than actual audience reach. Consider this analogy: A bookstore orders 10 times the quantity of books it actually intends to buy (or is capable of selling) in order to get a bulk discount. At the end of each month, it sends back 90 percent of the books it ordered. The store does this repeatedly, every month, until eventually the book supplier cuts it off.
In this example, the books represent reach, the book supplier represents the publisher, and the bookstore represents the network. The number of books ordered represents the network's "potential reach." The books not returned represent the bookstore's "actual reach."
At some point in the not too distant past there was talk about introducing the concept of "potential reach" into third-party ad network rankings. Rather than assigning a single lump sum reach figure (employed to date) to each network, the figure would instead be segmented as actual reach (or reach that a network actually serviced) and potential reach (the reach that a network had the opportunity to service).
Currently, reach is assigned to an ad network if it has received exposure to an ad impression at some point along the sequence of display. The problem with using this as the criteria for reach assignment is that the mere exposure to an ad impression does not provide any guarantee that the network exposed actually placed a paid advertisement. It only proves that the network had the "potential" to place a paid advertisement.
Why this matters
- Looks can be deceiving: You may conclude that a network (as above) is much larger than it actually is if you interpret its audience reach at face value.
- Instability: Networks (which amass reach as above) can be subject to substantial fluctuations in supply since they are not contributing significantly to publishers' bottom lines and are therefore vulnerable to displacement. Networks in this category may let you down on that media plan signed six months ago when they realize that they no longer have the inventory to fulfill it because their publishers have moved on.
Ask, "How much of a network's reach is fulfilled and how much is passed on?"
If your network does more passing than filling, its supply (and therefore your media bookings) will be subject to potential instability.
Lesson 3: Why reach diversity matters
Sometimes an old adage says it best: "Don't put all of your eggs in one basket." If you drop it, then you lose them all.
There is no shortage of networks that tout widespread horizontal diversification: 1,000 sites, 5,000 sites, 10,000 sites... but in many cases, the majority of sites that make up these networks are inactive or offer negligible contributions to overall inventory and reach.
If the majority of a network's reach is derived from a handful of large portals and/or large social networking sites (which is true in more cases than you might think!), and something disrupts that network's supply dynamic (e.g., you decide to block a major portal from the network's site list or eliminate all sites containing UGC), clients with bookings on that network may experience massive variations in campaign delivery.
Why this matters
- High vulnerability: Faced with the potential of under-delivery, networks in this situation may feel compelled to quietly "fill" your buy with un-optimized third-party inventory -- just to uphold their booking commitments.
- Higher probability of reach duplication: You may be buying inventory from the same large portals that your key networks are primarily composed of.
If your plan includes networks that lack reach diversity, you're at greater risk of encountering a disruption to your campaign's delivery at some point in time.
Ask, "How diversified is the network's supply of impressions and reach?
- What percent of the network's impressions come from the web's three largest portals?
- What percent of the network's impressions come from social media properties?
- How many of the network's "sites" actively contribute impressions (and reach) to the network?
In summary and to answer the ever-popular question, "Why would I buy from a lower reach network?" I leave you with the following:
Reach is a useful tool when you look beyond the figure itself and analyze each network's individual sourcing techniques and composition. What you find may surprise you. In some cases, high reach may also indicate high risk. In other cases, low reach may indicate high quality. You need to know how to spot the fundamental differences in reach composition that make each network unique, and to do so, you need to know what questions to ask. Ultimately it's your clients' budgets on the line, and your reputation.