Jay Sears Q & A with Jeff Arena, Senior Product Manager, Yahoo!


Jay-Sears As part of the run-up to the Context Revenue Strategies conference next week (on Thursday, October 5th only and FKA AdSpace at last Spring’s adTech in San Francisco), Jay Sears, EVP at ContextWeb, Inc will be running some terrific Q&A with some of the panelists from the CRS kick-off panel “The State of Content Advertising: The Players, The Options, The Best Practices” at 9am on Thursday, November 5th.

To kick it off, here is a Q&A between panelist Jeff Arena, Senior Product Manager at imagesYahoo! Content Match and moderator Jay Sears, EVP at ContextWeb, Inc and the ADSDAQ Exchange.

Your Name: Jeffrey D. Arena

Your Job: Senior Product Manager, Yahoo! Content Match. Myjob is primarily focused on defining and executing the global product vision for Yahoo!’s Content Match contextual advertising platform.  In this capacity, I oversee the global Yahoo! roadmap for contextual PPC text ad serving platforms that support Yahoo!’s Content Match text ad network.

Jeff Arena Yahoo

Jeff Arena, Yahoo!

Your Company: Yahoo! I work specifically within the Sponsored Search Advertising unit of the Advertising Product Groups (APG) at Yahoo! Sponsored Search Advertising as a group contains both of our global text ad products, specifically our Sponsored Search and Content Match products.  Our primary customers are Sponsored Search and Contextual advertisers, advertising agencies, and advertising tool providers.

It used to be contextual advertising was the “step child” of search, living in its shadow. It finally seems to be coming into its own. Do you agree or disagree and what are the macro forces contributing to this shift?

I do agree with the premise, which is that contextual advertising has long been an afterthought in a larger PPC, textual advertising campaign.  However, a combination of factors is fundamentally changing the relationship between Search and Contextual.  Specifically, these factors include:

  • Decelerating growth in algorithmic search query volume is causing advertisers to look elsewhere for additional volume (at some point, search growth will be bounded by growth in overall internet adoption)
  • Maturing reporting and distribution controls specific to Contextual advertising (and not Search) are allowing advertisers to unlock pockets of value within these large ad networks

Why does content and context matter?

Delivering ads alongside related content (e.g. in-context) allows an advertiser to get exposure to an audience he/she believes to be a target for their products.  So, content and context matter as a proxy for delivering audiences directly to advertisers.  For example, if I’m selling sporting goods to a college-aged audience, I might choose to contextually target fantasy sports sites for my advertising, as these sites tend to aggregate the same audiences as my product.  Over time, this disconnect between buying ads by context and buying ads by audience will close, as new and unique targeting capabilities emerge to allow advertisers to buy these audiences more holistically and more directly.

How does behavioral and demographic targeting tie in with content and context? Or does it? Mutually exclusive or best used together?

Building off of my previous answer, behavioral and demographic targeting are targeting methods that allow an advertiser to buy audiences more directly, as opposed to contextual targeting, which is a more indirect method of buying audiences.  I do not believe these targeting methods are inherently mutually exclusive, as there are a number of ways they could be creatively used together to access a very specific audience.  That said, if I were a contextual advertiser just starting out with behavioral or demographic targeting, I’d likely be testing the performance independent of other targeting methods before I introduced the complexity of combining multiple targeting types together.

The Long Tail and media fragmentation. More than 80% of Internet sessions start with search-the advertiser’s customer is now everywhere. Adsense has one million publishers carrying its ad tags. How do you compete in the Long Tail and against an installed based such as Adsense?

As an Advertiser, unfortunately, this dispersal of audience to the far corners of the web increases the friction in the buying process.  Since users are visiting increasingly diverse destinations on the web, advertisers are forced to deal with such audience fragmentation and chase those users across multiple ad network and advertising tactics.  Clearly, someone will benefit from the consolidation of such fragmentation at some point, but such advertiser complexity is an ongoing reality in the short-medium term.

Search has benefited enormously from last click attribution. More recently, Microsoft has published Atlas Institute research on engagement mapping and more advertisers are considering multiple attribution protocol when determining media mix. What are some of the macro “forcing functions” you see behind multiple attribution models and how will this benefit contextual advertising?

The general convergence of display advertising and search advertising, as well as the maturation of the performance display market, are both drivers for better attribution modeling.  When advertising online, large online marketers are typically spreading their budges across a number of buckets, such as guaranteed displayed, non-guaranteed display, search, contextual, etc.  Search and Contextual advertising, by virtue of being sold CPC to advertisers, have benefited disproportionally when being measured, from an overall ROI perspective, against non-CPC ad buys (e.g. guaranteed display sold as CPM and non-guaranteed display sold as CPM).  There is inherently too much friction in the advertiser processes associated with determining ROI across various online advertising tactics and such friction needs to be eliminated from the system as search and display advertising converge on the buy side.

That said, contextual advertising will not necessarily benefit from multiple attribution modeling.  Today, search and contextual ads that are clicked on by a user receive 100% of the credit for driving that click.  Distributing the value of that click across other ad buys will inherently lower ROI for contextual advertisers.  The devil is in the details, but such a paradigm change will have winners and losers, and contextual advertising is not a guaranteed winner.

Site targeting. When you move into the Long Tail (or even past the top 1,000 or 2,000 publishers), can site targeting deliver scalable solutions to advertisers? Is content a better answer because it is a common currency across all web pages?

I see Site Targeting as an extremely valuable tool in a successful contextual advertising optimization process.  Advertisers are not at all concerned with the scalability of site targeting in the context of a large ad network.  Ad Networks are the ones who are scared of this targeting technology because it threatens the subsidy they currently collect on low-quality traffic at the expense of high-quality traffic.  Advertisers should see these “scalability” concerns for what they really are and not second-guess the value and criticality of site targeting in managing a successful contextual advertising campaign.

Dynamic content. Web pages change constantly. How important is real-time – real time valuation, allocation, optimization?  Many folks who talk about real-time talk about “audience aggregation” and re-targeting, but how important is content as one element of a “real-time” decision?

As with any targeting, having a “real-time” understanding of the context against which you’re targeting ads is essential to long-term success.  But, by that same token, having a “real-time” understanding of your user, including where that user has recently been on the web or what things that user has recently been searching for, are similarly critical to long-term success.  That said, these are not necessarily new problems that have emerged alongside the “real-time” web 2.0 meme.  So, to a certain extent, both the newness and criticality of these concepts are a bit overblown, in my opinion.

Keyword vs. category targeting. Keywords are the holy grail of search. But are keywords effective in content targeting? Are they a destructive vestige of search-too granular or sometimes out of context to be impactful for content targeting? Is category targeting the answer?

Keywords have a limited value in the world of contextual advertising simply because very few parties attribute clicks and conversions to individual keywords.  Optimizing your keyword lists for a contextual campaign is fraught with trial and error, as advertisers cannot determine which keywords are having a positive influence on their ad group versus those keywords that are not.  Keywords would be more valuable for contextual advertising if keyword-level reporting and attribution were available for purposes of optimization.

However, keyword-level reporting and attribution are inherently problematic in the contextual advertising space as no single keyword is responsible for driving a particular ad-to-page match.  In this sense, categories are a more appropriate concept for purposes of reporting and attribution as many contextual matching technologies internally leverage categorization for determining contextual advertising matches.  That said, category-level targeting and reporting, although easier from the perspective of new online advertisers, does indeed offer less control in cases or both positive and negative keyword targeting.

So, clearly there are trade offs here when any ad network is considering this question in regards to their advertiser offerings.  In general, I think you should expect to see keywords used in contextual matching for some time.  Meanwhile, I would expect a number of distinct targeting capabilities to be added to standard contextual advertising offerings.  Among the candidates for such new targeting are: behavioral targeting, demographic targeting, user re-targeting, and category targeting.

Pricing Models. CPM. CPC. CPA. Cost Per Whatever-engagement, order-Cost Per Flowbee. Is this the direction we are headed? Good or bad?

I think these concepts are valuable as applied today, but the well established offline concepts of reach and frequency are not going away and will likely finder broader application within online advertising over time.  In the end, advertisers need a single, consolidated mechanism for evaluating the success or failure of marketing efforts and a single, high-level approach for allocating marketing resources across advertising tactics, both online and offline.

Can you sell content ads alongside search ads-1. With the same value proposition? And 2. To the same SEM buyer? Or is it more sensible to sell to agencies?

Absolutely!  Contextually matched ads can be and, in fact, are sold to the same advertisers with the same goals and with the same value proposition.  Whether or not one’s product can deliver on those promises is a different matter altogether.  That said, it is completely reasonable to assume a properly priced contextual advertising click or conversion can deliver the same value as a properly priced search advertising click or conversion.  The key, obviously, is recognizing these are two different methods of advertising that have unique characteristics in terms of end-user response and then pricing the products differently, to account for inherent performance variations.

Tell us about you.

What did you do last Saturday?

My wife was out of town last weekend.  So, I spent several hours on Y! Sports, tweaking and tuning my fantasy football team, while watching several college football games simultaneously!  That agenda will quickly be replaced when she returns.

What’s the best conference you attended in the last two years (besides AdTech and CRS, of course)?

I always enjoy Search Engine Strategies, San Jose, as there are typically some good sessions related to contextual advertising.  That said, I’ve seen more growth in the publisher-side content on contextual than on the advertiser side, so I appreciate the fair-and-balanced approach that CRS has taken.

If you could be appointed to any position in a US Presidential cabinet post, what position would it be and why?

Secretary of State – because I find diplomacy a more interesting and fruitful potential lifetime endeavor as compared with politics at almost any level.  Oh, and all the free travel would be pretty cool, as well.

Your LinkedIn profile: http://www.linkedin.com/in/jeffarena

Your Twitter account: http://twitter.com/jeffarena

Your Company’s Twitter account: http://twitter.com/yahoo

Thanks, Jeff!

Jeff’s Bio:

Jeff Arena is a Senior Product Manager at Yahoo!, focused on defining and executing the global product vision for Yahoo!’s Content Match contextual advertising platform. In this capacity, he oversees the global Yahoo! roadmap for contextual PPC text ad serving platforms that support Yahoo!’s Content Match network.

Before joining Yahoo! two years ago, Jeff was a Senior Product Manager for Ad Serving at LookSmart, Ltd., and also held a Product role at Neverfail, Inc. Complementing his Product roles, Jeff has held product development and IT consulting roles for Fortune 500 Financial Services companies while a Manager at Accenture, as well as a Software Engineering position, developing systems management software for Dell. Jeff graduated from The College of Engineering at Cornell University with a degree in Electrical Engineering.

About CRS

Produced by Marc Phillips and David Rodnitzky in conjunction with adTech, CRS is a “conference within a conference” is 100% focused on contextual advertising-an area that has needed its own conference for some time. Please join us for the CRS kick-off panel “The State of Content Advertising: The Players, The Options, The Best Practices” at 9am on Thursday, November 5th.

Our panel looks like:

The State of Content Advertising: The Players, The Options, The Best Practices

Where is contextual advertising going? What’s hot, what’s not? What can you do today to make your content buys massively profitable? This roundtable discussion features some of the top minds in contextual advertising sharing their insight on the state of the industry now, and in the future.

Jay Sears, Executive VP, Strategic Products and Business Development, ContextWeb, Inc./ADSDAQ Exchange

James Colborn, Director, Microsoft Advertising, Microsoft

Oded Itzhak, Founder and CEO, AdSide

Brett Brewer, President, AdKnowledge

Geri Guillermo, Director of Sales, BidPlace Pro, Sponsored Listings and AOL Search, AOL Advertising

Jeff Arena, Senior Product Manager, Yahoo!

Rajas Moonka, Group Business Product Manager, Google, Inc.

Special offer: Register today and receive $100 discount (Promo Code: CRSNY91) for the upcoming Content Revenue Strategies @ ad:tech NY.


Tags: , , ,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: