The Inventory Constraint

by Apr 22, 2005, 1:10 PM

• Comment
• Recommend (1)
• Tweet
• 

Buying online media has changed considerably over the past two years. It's no longer a buyer's market. We find ourselves at a time when online marketing dollars are increasing dramatically while agencies are not paying close enough attention to the resulting media waste. Demand for placements on well branded sites has outpaced supply, leading to the dual effect of higher CPMs and sold-out inventory (sometimes months in advance). Call up Yahoo! and ask for their home page box unit, and you will see what I mean. In response, I espouse three targeting tactics that when adopted widely will have a profound effect on leveling the uphill playing field that agencies now face.

1) Behavioral Targeting (BT)
BT has been heralded by many as the next big thing in online advertising. Some vendors sell this capability to clients as super enhanced targeting that can achieve search-like conversion rates and attract customers with the highest lifetime values. However, this selling strategy risks not attaining such lofty goals, thus squashing a client's taste for future testing.

What the industry must embrace is that while BT does have high return on investment (ROI) potential, its greatest immediate value might be increased inventory. As an example, let's say an agency wants to advertise on Yahoo! Finance to reach online traders, but finds most of the placements are already sold out because of the vertical's highly competitive nature. BT allows that agency to reach those same users on Yahoo! Mail because their profiles indicate that they check stock quotes on Yahoo! Finance several times a day. BT breaks the supply limitation of inventory by placement and opens the way for increased inventory, this time by behavior. Adoption of behavioral targeting across the industry has been slow and sporadic, but it is essential to creating further opportunities to reach high-value users.

2) Frequency Capping
Media agencies and clients likely do not need as much inventory as they think to hit efficiency and volume goals. Looking at frequency distributions (how many people receive one exposure versus two, three, etc.) and at the actual optimum exposure level, most online campaigns have somewhere between 10 to 40 percent media waste. That's because a small number of heavy users absorb a disproportionately high number of impressions. One person might take up 300 to 500 impressions over the course of a month. This is good for publishers, but bad for advertisers.

By asking publishers to place a frequency cap on a client's campaign, planners can achieve one of two results. For a site with expansive reach, the cap essentially becomes a re-distribution strategy, giving the impressions to new reach instead of wasting them on heavy users. For a small, well branded site where inventory is tight and CPMs high, the cap reduces your demand for quantity as smaller amounts of purchased inventory, served at the right frequencies, work much more efficiently.

3) Dayparting
For many campaigns, there is little need to buy all 24 hours in a day. User behavior tends to change from daytime searching to nighttime surfing. Although not always the rule, this difference creates higher conversion rates during the day but higher click-through rates at night. Depending on a client's goals, this contrast could lead one agency to request inventory only during nighttime hours while another buys daytime ones. The important thing is to understand which dayparts have a higher probability of achieving your goal and buy on that to reduce waste.

As more clients shift ever growing percentages of media budgets to online, it is incumbent upon media agencies to do everything in their power to slow the economic transition occurring between buyer and seller. Agencies as a whole must take advantage of every targeting tool at their disposal or face a future industry marked by drastic diminishing returns as inventory supply shrinks and prices rise.

• Comment
• Recommend (1)
• Tweet
• 

Be the first to comment on "The Inventory Constraint"

Leave a Comment