Quantcast

Ad-Based Sites and Content Networks

Get the WebProNews Newsletter:


[ Business]

Content Networks are the primary source of revenue for most Ad-Based sites. Measuring to optimize revenue from these sources is an interesting set of problems – but even more interesting are cases where a site blends revenue from multiple sources. Most common among these are sites that sell banner ads directly as well as Content-Network placements.

If a site is fully dependent on content-match ads, the only way a site can drive revenue is by losing the visitor to a click – which makes for a unique set of analytic problems. In this situation, the analyst needs to be able to measure a couple of key behaviors – visitors who click out on Ad-Sense ads, visitors who return same session after clicking out (big win here), and subsequent return behavior for visitors who click out and don’t return in the same session.

The analysis in the middle – visitors who click out and return in the same session – is often poorly supported by the default implementation of many analytic solutions. To capture this, it’s often necessary to capture the external link click and record a special event or page view. This, by the way, is quite a useful thing to do in general – since it provides a convenient pathing waypoint for the analyst to study. Since most of the key analyses for this type of site will focus on where and when a visitor clicks out, having the click-out as a page occurrence is quite useful.

Using this tagging technique, the analyst can reasonably answer key questions like: what pages are the best for driving click-outs, when in a session does a visitor click out and does the timing/location of a visitor click-out effect subsequent return behavior?

These can all be difficult types of analysis, but they pale in comparison to the problems that are faced by sites that blend content network ads and impression based banner advertising (or, for that matter, impression based site targeting). When a site mixes these two types, then each click-out is usually associated with lost revenue from impressions. So every revenue gain is offset by some level of revenue lost. To effectively analyze this type of site, you have to understand three separate interacting factors:

    1. How likely a visitor is to rejoin the session after a click-out

    2. How many pages a visitor would likely have viewed subsequent to the click-out if it hadn’t occurred

    3. How the visitors click-out changes subsequent return performance

The first situation is the best-case for a site – where the click-out is followed by a session re-join. The site gets the click revenue and doesn’t sacrifice any (or minimal) impression value. For point two, the analyst needs to understand how much impression value is being lost from a click-out. The three likely factors driving this are the area of the site being viewed, how deep in the session the visitor is, and the type of visitor based on previous behavior or finding cues. Building a model to predict impression value/loss is probably the single most complex task an analyst of ad-based sites will face – and yet it’s essential to understanding this and many other problems on a publishing site.

The third point is easier to deal with – and also quite essential. A click-out from a site will probably have a measurable effect on return rates. It’s important not to be fooled by epiphenomena here – regular visitors may account for most click outs. Which doesn’t make click-outs causal of regular visits. The key here is to analyze first time behaviors holding total pages-viewed constant and then comparing return rates over some modest period of time.

Only by careful analysis of the likely short-term and long-term effects of click-outs can an ad-based site make an intelligent decision about where, when, how often and how prominent drives to content-match ads should be. It’s easy to count the revenue from such ads – much harder to measure the potential loss. This may be especially true for sites trying to aggressively build traffic. Building traffic and monetizing it are two fundamentally different goals – the focus of which the business as a whole must decide and the analyst be responsive to.

One final thought for sites that are mixing impression-based (especially sponsorship) advertising with content-match ads. By capturing the click-out, you can also capture the sites you’re sending traffic to. This is more complicated since you have to parse the link from the served ads (no picnic) – but the information is potentially very important. If you find that a small set of advertisers are getting lots of your click-outs, these can become a natural sales opportunity for impression and sponsor based advertising. Most sites will never be able to remove the content-network middleman when it comes to aggregating buys from thousands of smaller players. But there is every incentive for ad-based sites to remove the middle-men when it comes to negotiating directly with your most important advertisers. This, at least, is a part of the business no publisher need ever give up. But, in this world of over-reliance on the search engines, many sites don’t even know who their largest advertisers are!

Comment

Tag:

Add to Del.icio.us | Digg | Yahoo! My Web | Furl

Bookmark WebProNews:

Gary Angel is the author of the “SEMAngel blog – Web Analytics and Search Engine Marketing practices and perspectives from a 10-year experienced guru.

Ad-Based Sites and Content Networks
Comments Off


Top Rated White Papers and Resources

Comments are closed.

  • Join for Access to Our Exclusive Web Tools
  • Sidebar Top
  • Sidebar Middle
  • Sign Up For The Free Newsletter
  • Sidebar Bottom