Author Archives: dwintz1

Originally published 8/10/07

How does an icon in traditional media make the transition from an outsourced ad operations solution to in-house ownership of the tools and processes needed to facilitate online advertising?

This is a story about how a logical, step-by-step roadmap can ease that transition. And in this case, how the process became relatively painless and ultimately successful.

This is one of those rare instances where the names are NOT changed to protect the innocent. So let me introduce you to Joe Friend, ad operations manager of Philly.com, the flagship internet property of the 150-year-old Philadelphia Inquirer.

A change in ownership of the newspaper created a situation where ad operations, including ad serving, contract management and all associated workflow, was being brought in-house on a six- week timetable. As a former analyst at the newspaper’s technical help desk, Joe was approached with the opportunity to manage the transition and the associated ad operations department.

As in most circumstances, the stage was set with a cast of players who had varying degrees of experience. Joe was experienced in internet technology, but not so much when it came to ad serving and operations. Philly.com had never managed its own ad server or inventory, but it did work with an outsourced operations group that handled all ads. In six weeks, the company needed to deploy a new ad server across the site, migrate hundreds of campaigns from their previous operations provider, provide a roadmap for process and workflow for the new internal department and train sales and traffickers.

The project plan

The first step was to create an internal project plan with benchmark dates that highlighted the critical tasks that needed to be accomplished. Although an ad serving provider may furnish a high- level schedule, it’s really in the best interest of the publisher to take that into their own hands. After all, it’s your site, your schedule and your job at stake. You own it.

So “let’s break it down,” to borrow a phrase from the music business. By that I mean, let’s break down this project into the major tasks needed to accomplish the transition. Each of these tasks needs an owner, a start date and an end date. In the case of Philly.com, here was the basic game plan:

Organization

• Create the project plan
• Create organization chart and define roles/responsibilities
• Develop a process workflow that highlights how to take an order from “quote to cash”
• Present project plan and workflow to sales and finance
• Gain agreement on the business rules that will drive the ad operations process

Define the Site Structure:

• Define the sites and sections that define the web properties and the types of packages that sales will sell
• Create the ad tags
• Technical call with ad server to discuss type of tags needed at Philly.com
• Build and distribute ad tags to web development group
• Test tags in QA environment
• Move tags to production server
• Create test matrix to verify ad units and placements and work

Administrative:

• Create users and set levels of permission in the ad server and contract management application
• Create the ad products in both applications
• Create any checkpoints or gate-keeping function

Contract Management:

• Train operations on contract management application
• Upload agencies, advertisers and contacts
• QA the site structure and make sure it syncs between ad server
• Load terms and conditions
• Load rate card
• Test billing output and discuss integration with current financial applications

Campaign Migration:

• Identify all ad campaigns that cross over from the old to new ad server
• Ensure that legacy campaign reports are archived
• Retrieve all creatives and clickthrough URLs from old ad server
• Calculate the impressions that need to be input into the new ad server in order to fulfill the balance of the ad campaigns
• Book all the currently running campaigns into the new ad server
• Create a communications plan for current advertisers advising them of the changeover and how transitional campaign reports will be delivered

Live Date:

• Coordinate launch support
• Create an escalation plan with key resources clearly identified
• Develop a testing matrix for post launch QA
• Monitor campaigns to track expected delivery and ensure they are on pace to fulfill contract
• Monitor performance of web pages and ads

Creating a project plan helps keep things organized. Without it, all of the above tasks can seem daunting and overwhelming. Even worse, without this type of plan and the clear owners and end dates, many of the items above can get dropped. Finally, anyone else in the organization can see a clear picture of all the work that needs to get done, it’s the best answer to the question: “What do these guys do all day?”

So, how did it all work out? Pretty darn good. The best indication was the eerie quiet after launch. No advertisers or sales people asking, “Where’s my ad?” No ads obstructing editorial recaps of the latest Philadelphia Phillies games. No frustrated staff members submitting resignations. Joe had a project plan and a couple of partners to help define it and execute it; in this case DMW MediaWorks and DoubleClick.

Ad operations is not a perfect world. Post launch, perhaps the biggest challenges are associated with measuring and forecasting inventory. But ask any publisher and they’re pretty much in the same boat when it comes to inventory forecasting.

All in all, the Philadelphia story is a good one. My advice to those in similar situations is to break it down so you can plan, organize and execute effectively.

Originally published 12/21/05

When our sun grows ancient and cold, and the inhabitants of this planet leave for a place far more hospitable, chances are one of the last lights flickering will be shining on an ad operations person trying to wrest the last impression out of a campaign.

Exaggeration? Of course. But I guarantee you that on December 31, 2005 every online publisher will have staff who are engaged in that very same activity.

So for this holiday season and those ad operations people in the far-flung future — we salute you!

Consider this my open holiday card to all of you in our industry, a card that acknowledges all that we can be thankful for, a card that reviews some of the past year and asks what you would wish (editorially speaking) for in 2006.

The year in review, and the year ahead

When we’re in the midst of all the year end activity it’s easy to slip into the old cliché, “the more things change the more they stay the same.” But there are intriguing hints that next year may see some more dramatic trends. Next year, I think we’ll get some new toys in our ad ops stockings:

Inventory management is on everyone’s mind. At a recent IAB conference, the following question was posed to a room full of ad operations professionals: “How many of you make use of the inventory management tools provided by your ad serving company?” Not one person raised his or her hand!

And this is the year when many, many publishers are finding that the biggest challenge is… there is no inventory left!

So the need to be more accurate in inventory management has never been greater. Who is going to step up to the plate and deliver the flexible, site centric inventory management tool that makes us less reliant on spreadsheets and more confident in an intelligent solution? Watch out for a couple of new tools coming out in the marketplace and make a wish that they deliver.

Ad serving has advanced to the stage that it is so reliable we count on it the same way we count on seeing a page of content when we type in a URL. The main stalwarts of the ad serving business — and they have been around for the last few years — are thankfully still here. But we may also see the field get more crowded as some upstarts make inroads. We may even see some new players make an even bigger splash by entering the field. If so, 2006 could see a dramatic change in this landscape. I won’t name names (since I don’t aspire to reporting for the National Enquirer), but keep your eyes and ears open in 2006.

Video. If you are looking for entry into an area that will be in constant change for the next five years, get involved in video. Our entire landscape is going to be changed several times as we strive for consistency and standards in video ad delivery, reporting and pricing models. Then, get ready for it to change again as true convergence in broadcast and internet media begins.

Just when you thought internet media was safe because we finally agreed as an industry on the definition of a banner ad impression, get ready for the same dialogue on video. Frankly, I call it the holiday present of job security — we will need experienced ops people to figure this out over the next few years, and as we all know they (I should say “we”) are in short supply at the moment.

Speaking of jobs — the best gift for 2006 will be the excellent job market. But we have certainly paid our dues in the past. Who remembers the climate in 2000 through 2002? That pitiful person on the street corner in a raggedy coat, talking to themselves, glancing furtively with rapid, unfocused eye moments, gesturing aimlessly to nobody in particular? Why, that was probably one of our ad ops or ad sales brethren during the bust. Look at us now.

On the other hand, let’s not get too smug. In the interest of self preservation, I would urge you all to look out for bright, young protégés to mentor and get them involved in your business. The flip side of being in demand is that you will be overworked. At some point, no matter how highly you are valued, it won’t be enough to compensate for the long hours spent doing the jobs of two or more people.

There’s much to be thankful for.

Pragmatically speaking, it’s hard to stop and take the time to be thankful for anything in Q4 other than just getting out at the end of the day alive and with sanity intact. But if you take a step back then it’s clear that there truly is much to be thankful for, and I’ll help you out.

In order of importance:

Our family and friends. Father, mother, sister, brother, spouse, child. We all can count on at least one of these individuals for support and affection when all else fails. They won’t berate you for taking too long to test an ad tag, or blame you for late creative. Remember that between now and the New Year.

Our business. You are engaged in an industry whose rules are yet to be written. You are part of a future that pundits can only guess at. Chances are the next 10 years will be filled with invention that will challenge and stimulate you. You will look back in 30 years the same way your parents did at the early days of television or grandparents did with radio — and know that you helped create the future. You will not be bored.

The ad ops teammate who covered for you when you went on vacation earlier this year, or when you came down with the flu, or had family business to attend to. The teammate who perhaps mentored you, or showed you the ropes when you were new, or even referred you to a new job in a more pleasant (or at least more lucrative) environment.

The sales person who said “thanks” for your efforts at getting the campaign live in less than three days, or for the monitoring and optimitizing you did that led to a big renewal, or for wrestling that big rich media ad unit to the ground and getting it live in time for a key promotion.

Finally, I’d like to thank all of you for your comments and input this year in response to my articles. I can’t tell you how gratifying it is to receive the 10 to 15 emails after each article with comments like “we’re glad someone finally wrote about this” or “we were just discussing this topic at our staff meeting”. It’s kind of like cheap, anonymous group therapy, isn’t it? (Well, it is to me).

What do you wish for in 2006?

So now I’ll ask — what do you want to hear about in 2006? Oh, I have my own ideas. The advantages of a local versus ASP ad serving product. The implication of new targeting tools on ad operations like behavioral and bid management tools. But, what is on your mind that I haven’t yet covered? I’m interested. Drop me a line.

For reference, below are the topics and articles that I did cover in 2005. Keep the links handy if you want. Think of it as a refrigerator magnet for your desktop on ad ops matters.

Above all, have a great holiday season filled with family and friends, and complete with the knowledge that while your job may be stressful, it reflects well on your desire to stretch your intellect well beyond your comfort zone, which makes you a risk taker, an adventurer and a pioneer. Sounds good to me.

Best wishes,
Doug Wintz

Originally Published 07/03/08

Most of us in ad operations are used to creating “work arounds” as a means of solving our most pressing problems. We are frequently handicapped by applications that are limited in their flexibility. We tend to adopt the attitude that “this is the hand we’ve been dealt, so let’s try to make the best out of what we’ve got.”

In actuality, we should be defining exactly what we need in order to run our business more effectively and demand that our company support us in creating those solutions.

Nowhere does this dilemma apply more than to the challenge of inventory management. Ad servers offer a basic level of inventory forecasting, but they fail at creating solutions that are publisher specific. You may be a news, travel, weather or automotive site, and in each case the challenges associated with granular targeting of inventory, of spikes in activity that can’t be accounted for by ad servers or of calculating revenue share with affiliates call for spreadsheets and manual labor to create marginal solutions.

However, publishers can cast off the shackles of ad server dependency by creating their own inventory tools to address their specific needs. Using the ad server API, publishers can write custom programs that automate many of the most arduous tasks of inventory management.

Ad server API’s (application programming interfaces) are essentially gateways to the inner workings of the ad server. By using the API, you can access delivery data, campaign data, advertiser data, forecasting algorithms, specific targeting as defined by key values, etc. You can use the API to create middleware applications that may utilize a language like Java to extract the inventory data and deposit it into a database. The same application can also utilize a language like PERL to take that data and format it visually into the type of reports you need and deliver it to the ad operations group on a daily, weekly or monthly basis.

Below is a schematic which offers a top-line view of the interaction between the ad server and the middleware application that creates the reports:

What kind of reports can you create using this method? Just about any type of inventory report as long as the data is in the ad server. From there, it’s simply a matter of defining how you want to manipulate it. Here are some examples, all of which are currently in use by publishers.

Daily inventory snapshots

What cross section of inventory availability means the most to you on a daily basis? The “snapshot” approach compiles a customized matrix of the ad sizes, site placements and key values (think DMA, product attributes, etc) and deposits them on the desktop of the ad operations group. Based on the specifications of the publisher, the list of matching, booked and available impressions can be displayed over weeks, months, even a year. Seasonal trends can be accommodated by allowing the publisher to configure a table that takes into account adjustments due to specific events, traffic patterns or new site launches.

Automated query tool

Pulling inventory avails can be like pulling teeth, particularly for campaigns with multiple line items. Instead of working around this issue, why not create an in-house solution? A simple user interface can be constructed and used to query the ad server database, compiling a multi-line avails report. Further, the results of those queries could be emailed to users on completion, stored in a database so historical trends can be analyzed and re-used as a template for future availability pulls.

Revenue share reports

One of the most manual of tasks for publishers who have a network of affiliates is calculating revenue share. This is further complicated by the fact that revenue share can vary by affiliate, product and category of advertiser. If publishers apply the discipline of entering CPMs into the ad server, using the same methodology of accessing the API can yield a customized report that matches a table of revenue share percentages with the delivered campaigns by affiliate.

Gain control of your inventory

One of my musician friends uses the malapropism “What’s holding up the delay?” to describe the frustration of waiting for an event that never seems to take place. So I would ask publishers the same question. Why wait to take control of your inventory into your own hands? Most ad servers have an API, the data is there and waiting to be spun to your specifications. All it takes is the willingness to go from blind acceptance of your current limitations to creative problem solving on your own terms.

Originally Published 05/24/09

Mainstream publishers are leaping into the ad network business with increasing frequency. Why, you may ask? After all, isn’t it hard enough for publishers to fill the ad inventory on their own domains? Or have they secretly been selling out 100 percent of all their ad placements without telling anyone? If so, I hope they haven’t been taking TARP funds under false pretenses.

But seriously, there are some rational reasons for embarking on this strategy. First, creating a network of sites with similar content helps extend a publisher’s dominance in a particular category. Build a large enough network of affiliates and, presto, you are no longer the No. 25 site in MediaMetrix; you are No. 3. Impressionable media buyers will be impressed with your reach, and shareholders will be happy with your ranking.

Second, build a network now, and you’ll beat your competition to it and leave them in the dust down the road. And finally, to be fair and balanced, some publishers actually do find themselves selling out of highly targeted inventory in travel, health, and automotive, even if a fraction of run-of-site inventory actually gets sold.

For all these reasons, be they rational, sound business reasons or not, websites with a single dotcom domain are extending their reach by building networks. However, many of them underestimate the challenges associated with this model. Believe me, it is not for the faint-hearted or unprepared.

Here are some of the most unappreciated and underestimated challenges associated with running an ad network, especially if you are a publisher starting out as the new “owner-operator.”

• The affiliate websites in your network are actually your customers. Sure, you are good to your advertisers and have experience servicing them, but now add to that equation the fact that you have 150 network affiliates all needing varying levels of support and information. Make sure you have the staff to support that, or choose a solutions provider that does.
• Affiliates will need assistance in deploying your network ad tags on their site. Sure, you can sign 150 affiliates to your network, but can you actually get them to deploy your ad tags and on a timeline that suites your business model? Your help will be needed, and it will be technical in nature.
• As a network “owner-operator,” you will find that each affiliate will have separate business rules dictating which ads or ad categories you can and cannot show on their site. One affiliate wants to prevent you from selling to sports advertisers, because they have a standing exclusivity with a single firm. Another site wants to bar an entire list of specific advertisers. None of this type of control will scale unless it is automated, so your network solution better supply this.
• You will need to supply individual revenue reports showing how many impressions ran on each affiliate. Those impressions may be subject to a different revenue share depending on targeting and site placement of sales channel. Your affiliates will want to see them — and see them frequently. As a mainstream publisher, can your current ad server do that?
• Inventory management. If you’re lucky, your affiliates will cede control of 100 percent of their inventory, in which case, you will know where you stand in terms of sold and available inventory. But in many cases, you’ll get a portion of their inventory, sometimes on an ad hoc basis as you need it. In which case, knowing where you stand on inventory will be challenging at best. Make sure your existing ad operations group is looped into this function.
• Payments. Yes, your affiliate customers will actually want to be paid. Are you personally going to write out the checks and send them out? If not, then who will?

Your options for supporting a network are varied. Mainstream publishers can attempt to do it with their existing ad serving solution and internal support staff, but frankly they are doing it at their own peril. The danger here is that you will be complicating an already complex business model by relegating it to a series of internal spreadsheets and marginal customer support from your internal resources, which are probably already overtaxed.

Your best bet is to outsource support of your network to firms who are experienced and focused on that business model instead of using a do-it-yourself approach. Make sure you can get as close to an end to end solution as possible, ranging from deployment of tags, to customer support, trafficking and optimizing, inventory management, and billing/invoicing.

Expanding your business to a network model is not for the faint hearted, but given the right planning and resources, in can expand the reach and exposure of your core brand as a publisher.

Originally Published 08/13/05

Jon Stewart has described the internet as “a technology combining the credibility of anonymous hearsay with the excitement of typing.”

I’ll admit it. I’ve been trying to find a way to crowbar this favorite quote of mine into an article for months. I even toyed with the idea of going minimalist and just posting the quote as my sole contribution to the August column. But I didn’t think I could convince the editor that two lines of text and two pages of white space would be beneficial to our readers. Fortunately, there is a sound, rational reason to apply this quote to one of the most active, and most underdeveloped, trends in online advertising — video.

We are quickly leaving behind the era in which our medium is described, even flippantly, as “exciting as typing.” For those of us who have been in the business for a few years, the long-promised potential of video seems to have mushroomed almost overnight. Can you name one company at this juncture who has the cojones to say “video is not part of our strategy?”

Interestingly, the operations end of the business seems to have been caught napping when it comes to supplying an integrated application that combines video content management, a flexible, configurable media player, video ad serving that can be sequenced at will and back-end reporting that mirrors what we’ve all come to expect from the top tier ad delivery systems.

There are many vendors that solve a piece of this equation, but few, if any, who provide a packaged solution. As a result, these pieces need to be cobbled together from disparate service providers.

It’s like the bad old days of online advertising: no clear set of standards and no consistency in display of video advertisements. But that also means the medium is ripe for opportunity.

Here’s my wish list for consolidated video management system from a single service provider. (Sorry, but saying you have an API that allows other service providers to integrate with your application does not count.)

Manageability: A content management system that organizes and delivers both video programs and advertising to the player. Let me go to a library and pick the ads and the content from a well organized database.

Flexibility: The flexibility to sequence ads and content at will, by vertical. If I’m in Entertainment, perhaps I want to show an ad before each content clip and not worry so much about frequency capping. If I’m in Health, maybe I only want to show an ad every three content clips and impose a frequency cap as well. If I have a premier sponsor for any of these areas, I might just want to tailor the display of video ads to meet their requirements.

Customizability: A video player that can be customized to meet the needs of the publisher. This ranges from simply skinning the player, to running standard ad units in tandem with video, to configuring a play list “on the fly” without the need to undergo custom programming each time.

Targetability: Reporting and targeting of video ads that mirror the accepted standards from pure ad serving applications. By content section, daypart scheduling, geo, frequency capping. Don’t forget to throw in drop-off rates — how many people view the entire video versus only the first five seconds?

Oh, and can this service provider please also include the ability to serve standard and rich media ad units? Just like the big boys do.

Sure, it’s probably unreasonable to expect all of this from a single provider. Companies are prone to focus on a specialized aspect of media operations so they can excel in at least one thing. That’s good business practice.

But we do need to move away from the current environment of piecing a jigsaw puzzle together in order to serve up the most valuable, highest CPM inventory for any publisher.

Finally, there’s another reason to consider this topic carefully. The internet will drive the future of television. The tail is already starting to wag the dog. And in the future, when someone says “we’re live in 3, 2, 1….” a media application will start serving video content, clickable video ads, banners, ecommerce links and collecting a database of interested users.

Someone will have to power the media, and they’ll have to figure out how to make all the disparate pieces fit together.

Published 01/28/06

As I’ve mentioned in previous columns, the life of an ad operations staffer is filled with day-to-day tasks revolving around the launching of new campaigns, troubleshooting creatives, revising campaigns, managing inventory– you know the drill.

Today, however, I’d like you to consider that you know far more than your corporate colleagues think you do. After all, who else has more intimate knowledge of your sites’ performance for advertisers? This includes how ad units perform in various site sections, for specific types of advertisers and at what time of the day or week. And that’s just the tip of the iceberg.

Frequently, because of the tactical nature of our business this institutional knowledge is overlooked, or — even worse — left unsaid. It therefore becomes one of the best-kept secrets of ad performance on your site. How many times have you been at the receiving end of a new campaign and said to yourself, “Oh man, here we go again. I know from experience that this isn’t going to work.” And then you move on to the next campaign.

One of the keys to your personal success in 2006 may be your ability to translate some of these best- kept secrets into solutions that create new ad products or packages. In doing so, you could be the source for new revenue streams– a feat that goes far beyond simple trafficking of ads.

Challenging Content

Let’s take a look at an example of how your knowledge could be translated into new solutions for your company.

One of the internet’s great strengths is that it provides a gathering place for communities of users. Their value has been proven time and again. Just look at history, past and present. Geocities was acquired by Yahoo! years ago. MySpace was acquired by NewsCorp months ago. Why? Because massive numbers of like-minded users gather together in these communities, and with one acquisition a company can increase their audience significantly. Recently the phrase “social networking” has been used to describe similar sites.

From the ad ops perspective (that’s you), “communities” show a different face. Sites that are composed of user-generated content create uncertainty among many advertisers. What kind of content will their ads appear on? A bona fide travel site with information about Cancun? Or, a “Girls Gone Wild” wannabe site carefully crafted by a frat boy just back from Spring Break?

Even if advertisers are not deterred by this uncertainty, there is the matter of performance. From your seat in ad ops where you schedule ads and observe performance, you know that campaigns in a community setting have shown historically low click rates. Therefore, CPMs for community sites (and this goes for community, social networking, email, chat and message board sites) are at the lower end of the rate card scale.

Be the hero

So, given all the knowledge you have as a professional in ad operations, what could you do in this situation? The options are 1) ignore it and keep scheduling campaigns 2) be the hero and come up with some potential solutions.

Think about what you do know. For instance, you may have observed that the “community” type content you work with is composed of sections that are highly populated by people interested in travel, automotive and music. What if you could aggregate the best of that content and create front door sections to those content areas?

For instance, your publisher or director for community content could select from the best travel sites– whose content is consistently of high quality and perfect for any advertisers sponsorship. Supplement this with some data feeds with news on travel, auto and music. (This isn’t brain surgery by the way– a company called Mining Company started something similar to this in the 1990s, and it became About.com.)

Now, you’ve used your knowledge about the behavior of your sites’ content, and created viable sponsorship areas that are attractive and a safe haven for advertisers, with more engaging content and perhaps even better performance from a clickthrough standpoint.

Why didn’t the founders of your site figure this out? Maybe they were preoccupied with building the community destination. Maybe ad revenue was a secondary consideration. Maybe the site has been around for a long while and there is a reluctance to retrofit into a new model. Or maybe, just maybe, they don’t know what you know?

Summary

You’ve got more knowledge than you may be giving yourself credit for. And new ideas are not the sole domain of writers, editors and marketers. In fact, your unique perspective on the business might yield significant new revenue opportunities.

08/03/2007

I’m not one who gives in easily to hysteria. After all, if you work in ad operations, you eventually become desensitized to emotions of panic and horror; the same thing that happens when you watch too many violent movies. However, when the latest round of acquisitions occurred in our sector, leaving virtually no tier one independent ad serving companies left standing, the following dialogue came to mind:

Fire and brimstone coming down from the skies. Rivers and seas boiling. Forty years of darkness. Earthquakes, volcanoes…
The dead rising from the grave.
Human sacrifice, dogs and cats living together — mass hysteria

(Ghostbusters, 1984)

For those of you who have been off meditating in Tibet for the last two months, I’m of course referring to Atlas’ acquisition of Accipiter, both of whom were bought by MSN; Google’s acquisition of DoubleClick; WPP gobbling up 24/7 RealMedia; and Ad-Tech being snatched by AOL. Amazingly, this happened all within the space of about three weeks.

Is it really the end of the world as we know it? To gain some perspective beyond my own, I had a chance to talk to several publishers and agencies over the last few weeks.

Short term, the news generated all the intense response of one hand clapping. In other words, it was business as usual. In a way, that just makes sense. Publishers still have campaigns to serve and revenue goals to meet. Ad agencies have clients to service and budgets allocated to the distribution of online advertising. Leaderboards, skyscrapers and messaging units all need to be served.

Long term, however, it’s another story. There are several implications to the changes in this end of our business. They are as follows:

Privacy

No, not individual privacy (frankly, there is none so don’t post anything online you don’t want on the evening news). I’m speaking primarily of publisher data. Now that the ad servers are all owned by other publishers or ad agencies, they have a view into CPM and pricing data that can potentially give them competitive information on online sales trends for the automotive, or business, or entertainment sectors. Keeping this data secure and confidential — even in this world of new ownership — will be crucial to maintaining the trust that is an important part of the publisher’s relationship with the ad serving companies.

Network dominance

I don’t have an inside track to the strategic plans for any of these companies. But if it were me, I’d go to every publisher and say: “We’ll greatly reduce your ad serving fees (or give it to you for free) if you give us custody of your entire unsold inventory.” This would allow Google or MSN to grow their network reach tremendously. The revenue from that growth might just offset any losses due to decreased ad serving fees. The long-term implication of this would be to further commoditize the cost of ad serving, making it even more difficult for a new and independent ad serving firm to get a foothold on the business.

Where your bread is buttered

We all have a tendency to focus on the aspects of business that can most efficiently get us the most amount of money. This is just as true for a corporation as it is for an individual salesperson. If the new ad serving owners do indeed go after network dominance by increasing their focus on gobbling up more publisher impressions, where will that leave their agency clients? Will support of agency side ad serving products be put in a maintenance mode, while the other side of the business gets all the attention?

Fox in the henhouse?

Isn’t it ironic that the new parent companies that now own the top two agency-side ad serving solutions have business models that are competitive with the very agencies they service? For example, Google is intent on fostering its self service/exchange model and rolling it out into all media forms.

Where does that leave the ad agency? I’m not arguing for one model or the other here because the answer is somewhere in the middle. But the lines are getting blurred and, long term, this may have implication on how agencies develop the operations end of their business.

Standards

One thing that may be more convenient in this whole scenario is the issue of standards related to the methodology of counting impressions, the methods of targeting ad campaigns, and of pricing. After all, if there are only two humongous ad serving companies, there’s little variance in methodology left to debate. This might leave the IAB with little left to complain about. However, consider the alternate future of many ad agencies replicating what WPP has done in acquiring an in- house ad server from 24/7 RealMedia. What if all the agencies do that or developed their own solutions: are we then headed to the disintegration of standards and a regression into Babel?

To come full circle on all this: I will admit I am engaging in a little bit of sensationalist journalism. And to be frank, I had fun doing it! Short term, it’s likely nothing remarkable happens. Is this REALLY a cataclysmic event in which the earth will open up and swallow all ad operations departments in one horrendous moment? Naw. If anything, it may be just a steady erosion, the kind that happens when you build your house too close to the seaside. The degree of danger you’re in depends on how close your house is to the water. So, what’s to worry!?

Originally Published 05/11/05

Before you pass this article by in favor of something you believe is more exciting, like streaming video ads, wireless multi-player games or interactive TV, think back to the last corporate fire drill that occurred towards the end of a fiscal quarter. This is where the CEO asks “how much are we selling, are we going to make our numbers, do we have the inventory to fulfill campaigns, what is our sell- through rate, why do we have unsold inventory and what are we going to do about it?”

If you have ever been caught in the eye of this storm, you either wish you had people who could supply this information on a moment’s notice, or you patted yourself on the back for having the foresight to staff up with an inventory analyst.

This article is about inventory, the importance of managing it and the people who can help you do that. Along the way, I’ll cite some hair-raising real life stories on inventory “mis-management” that unfortunately, you may be able to relate to.

First, let’s take a step back and put the issue into perspective.

For publishers who depend on media dollars for their livelihood, selling is just part of the equation. From there, insertion orders are routed to an ad operations group and are scheduled into any one of several ad serving applications available on the market today. Most of the brand name ad servers have a function that enables sales and/or operations to get information on available inventory. Sales and operations folks know this drill all too well. Most would also agree that inventory management systems are still a “black box” that can only tell part of the story and are at best inaccurate. They can’t take into account the day-to-day idiosyncrasies of selling in today’s market with today’s products.

Many publishers rely on their ad trafficking staff to give them insight on inventory. The problem is, most of them are so busy with the day-to-day tactical chores of receiving orders, sending them back for corrections, testing ad tags, responding to customer requests to change campaigns, that the last thing they have time for is thinking about inventory.

The job of thinking about inventory belongs to an “inventory analyst.” Unfortunately, for many publishers, this is a head count they don’t consider, or think they can’t afford. But as online media gets more complex, they will find they can’t afford NOT to have a staff position for this individual.

Now, before your eyes glaze over, I’ll relate some real life adventures in inventory terror that could have been prevented by having an individual devoted to understanding this part of the revenue equation:

• A publisher finally gets the Packaged Goods client they’ve been dreaming of. They are interested in promoting a candy bar and want to use a rich media floating ad, displayed during the afternoon hours of 1 pm to 5 pm, to reach a youth audience just when they need an energy boost. The ad trafficker takes the order and is so busy, he just takes the order. He fails to notice that 1) the rich media ad takes a total of three impressions to load 2) there is a frequency cap of one per user per day. The campaign starts to run, and you discover that the 10 Million home page impressions you thought you had has now turned into 1 Million. The client is furious about under delivery. The salesperson is at your desk wanting an explanation. The make-good you need to run to fix this has a ripple effect that impacts every other client on your website.
• It’s the beginning of Q1 and everyone is just getting back to work. An order comes in from a client who says, “I’ll buy every pop-under you can give me for a $5 CPM. My budget is $400,000. Just send me the IO.” The order somehow gets signed and noted as booked revenue. Unfortunately, based on the number of uniques on the site and the frequency cap and the fact that there are seven other rich media campaigns scheduled during the quarter, there’s only $200,000 worth of pop-under inventory to serve. As the quarter goes by, everyone is happy with the booked revenue. As the end of the quarter approaches, management wonders why $400K of booked revenue isn’t being delivered and comes down on the traffic department who scheduled the ads and asks them “why aren’t you monitoring our inventory?”

The solution to avoiding these problems is to have an inventory analyst on staff who, on a daily basis, monitors the site to determine how campaign delivery is matching up to booked revenue; an individual who knows how to calculate available inventory for complex campaigns that involve frequency caps, rich media, demographic, behavioral and geo targeting. The inventory analyst should also be able to create trend reports that give the business insight into how their inventory is being utilized, and recommend pricing and product strategies to increase yield.

The airlines have spent billions of dollars creating reservation systems that help them manage their inventory. Since your livelihood depends on making sure that booked equals delivered inventory, shouldn’t you consider spending slightly less to make sure your staff can give you the ability to monitor and manage your business?

Over the decades of working and consulting in online and internet business, I’ve heard some pretty weird things coming from vendors. So whenever I come across behavior that just seems wrong, I try and make sure I file it in the “mistakes not to make” folder. Here are some things I know rub publishers the wrong way:

We’re Really Busy Right Now. I get it, who isn’t? But if I’m the customer (publisher) and I’m paying you (the vendor), WTF? If you’re too busy, maybe you’re understaffed which makes me wonder about your ability to scale. I get it, resources are limited across the board, but there are much better ways to handle this. How about, “I’ll have a schedule for this requested task to you by COB”. Then you can regroup, address bandwidth issues internally (and not in front of the client) and in responding, you’re providing them a solution

We’re Getting Too Many Emails. Literally, I had a vendor complain to keep them off email strings because they were getting too many. Boo-hoo. It’s the price for having customers. And who knows, you might actually learn something!

Here’s a Link. This is a classic response to a request for more information. What does it say about the vendor? They can’t take the time to answer you in their own words. Or they don’t understand the topic and hope you’ll figure it out for yourself. Not a satisfying experience to a customer.

What’s Your Business Reason for Doing This?  This is really a nuance, but most often the intent behind the statement is “I don’t think your process is valid” or “I don’t want to do it and hope that I can talk you out of it”.  Again, a nuance, but a much better way to getting the question out is to ask for an example to illustrate the request, or a “use case”. I know, I know, splitting hairs here, but words matter. Well, at least they did as of 2012.

I’ll Have to Get Back to You. I don’t object to hearing this once in a meeting, but if I heard it 3 times in the same meeting, which I have, I would think that this vendor is 1) not prepared for the meeting and 2) doesn’t know their own product. If the purpose of the meeting is to inform, be prepared to do so.

Is This Project Worth Our Time? Ahh, one of my favorites. During a new business pitch, asking if the project is worth their time because they are busy and have to prioritize their work. Honestly, part of the job is called business development. Sure, you might not get the business today, but you might tomorrow, or next year. And we should be building our customer relationships over time, not because we need to close out the quarter.

In closing, I’d just like to say that we should be thankful for being engaged in a client’s business, and not the other way around.