What does digital need in order to get to parity with broadcast reach and frequency metrics? Here are four areas where we come up short.

Web advertising has the best targetability of any media vehicle out there, yet we can't seem to get it together from a reach and frequency perspective.

When various champions of the R/F/GRP model pushed hard to bring this old media model to the web, I kicked and screamed. It's a poor model for evaluating internet delivery, but I was assured that many advertisers need to see their digital delivery expressed in terms of reach, frequency and GRPs in order to see things on an apples-to-apples basis. It's especially the case when advertisers are into things like media mix modeling.

I kicked and screamed, but at the time the digital media industry was in no position to push its own model or force traditional media to reevaluate its established metrics. So I stopped kicking and screaming for a while.

Since then, R/F/GRP tools have become established and have gone through several iterations and revisions. But we're still missing a few important pieces, and it's making me feel as if I should have kicked harder and screamed louder. Here are some of the areas I think the measurement companies need to improve on:

1. Getting beyond basic age and gender: It's fairly easy to evaluate a web plan against basic age/gender demos like A18-49 or W25-34. Start adding household income, education or other qualifiers, though, and the numbers require hand calculation. If the audience is niche enough, the numbers will completely fall apart. While it's true that the more niche the audience, the less likely television networks are to guarantee the buy against that niche -- that really doesn't matter.  Advertisers are used to evaluating buys against demos they don't necessarily buy against. So we need to fix this.
   
   
2. Making it easy to see reach cume over different time periods: I'd eventually like to see a delivery report that can use a drop-down menu to toggle between weekly, monthly and quarterly deliveries. TV buyers and traditional planners look at things this way. There's a lot of legwork involved on the digital side in delivering on the same expectation, and often hand calculation is required.
   
   
3. Accounting for frequency capping: Often, digital media buys on different sites have different methods for capping frequency. We buy this way in order to extend reach and keep our prospects from being hammered over the head 30 times with the same display ad. Some sites might cap on a per-session basis. Others over the course of a day, a week or a month. I have yet to see a feature in the R/F tools that handle this gracefully.
   
   
4. Accounting for targeting: Given the degree to which we're putting things like behavioral targeting to work for our clients, we ought to have a feature that makes it easier to evaluate targeted campaigns on an R/F/GRP basis. Right now, evaluating most of those targeted campaigns requires some fudging and educated guessing.

Sit a VP of marketing or a brand manager down at the bar, start to imbibe, and eventually they'll tell you that digital doesn't get a larger share of advertising budgets because of the uncertainty of its delivery. Whereas, an advertiser can call an agency, ask to see what a $5 million TV buy looks like, and get something back from the agency in short order that details the deliveries of a prototypical plan.

We still can't do that in digital very well. When advertisers call the TV buyers, they get a concise delivery report. When they call the digital folks, they get a turning hourglass.

I'm sure Dave Smith from Mediasmith will add to this in comments, as reach and frequency are his pet issue. There are other problems here we need to address. But I'd like to see the digital media industry at least get to parity with broadcast when it comes to R/F. We can't afford delays or shortcomings if we want to continue to grow digital's share of the ad budget.

Tom Hespos is the president of Underscore Marketing and blogs at Hespos.com.