Jeremy Dunck Staff

One nice thing about a tech whose primary function is communication: only one person in a group needs to understand how to use it in order for the benefits of access be significantly transferred to all in the group.

2 months ago ( Link to this comment | Suggest removal )

Brett Hoerner Staff

djkeith, Mike is out of the office for a few days, I'm sure he'll get back to you.

In general though, sites that sell Advertising have to trust third party sites because people want "apple to apple" comparisons.

2 months ago ( Link to this comment | Suggest removal )

Brett Hoerner Staff

I don't think we have a lack of faith in the demographics. I think they're pretty spot on.

Mike's post is about how we get a bum rap by people who (wrongly) assume that because we're online-only, we only attract youngsters. And that is (as he proved above) not the case.

2 months ago ( Link to this comment | Suggest removal )

Gwen DuVal Staff

Actually, we have updated numbers coming that show in comparison to the norm on the internet, we reach more users in the 24-34 range than would be expected, and more users in the 55-64 range than would be expected. Both age groups that spend more time going out and doing things, versus the age group in between those two ranges, that use all their discretionary income paying for children.

So we really reach the people that restaurants, events, venues, and other want to see coming in their front doors...and yes I do sell advertising for the site!

2 months ago ( Link to this comment | Suggest removal )

Brett Hoerner Staff

But we do give them access. :)

I don't know how technical you want to get, but if you check the page source for any page on our site you'll see that we include scripts from sites like Omniture, Google Analytics, and Quantcast which report back information so each of them can track traffic.

I'm slightly oversimplifying, of course, but that's the general non-geeky gist.

2 months ago ( Link to this comment | Suggest removal )

Gwen DuVal Staff

okay, I give an overcomplicated answer and the voodoo prince of all that is complicated gives the simplified answer...that ain't right!

2 months ago ( Link to this comment | Suggest removal )

Mike Orren Staff

A little knowledge can be a dangerous thing, djkeith-- I'm late to this thread as I'm just back from vacation. Brett and Gwen have answered well, but I'll add a little more color on a few points.

"who ever said you were just for the kids? that sounds made up."

I got a good laugh on that one. Back when we were pitching for capital, that was one of the most common objections. It comes up in virtually any sales call that isn't to a bar or nightclub. And I know of at least two specific ad deals that we've had "in the bag" that got killed over that very issue. One was a large sponsorship from a large sandwich chain where the marketing guy had all but signed the contract and then the president of the company nixed it because he "didn't believe anyone over the age of 25 was on a site like this."

...

As to sources for numbers: Web metrics are the messiest thing imaginable. Apples-to-apples comparisons are nearly impossible (particularly for local) because no two players do or count the same way. And that's not all shiftiness either. At various times, we've used:

• Omniture (industry leader in traffic measurement) is our current internal metrics provider
• HBX (another industry leader who was recently acquired by Omniture)
• Google Analytics
• LiveStats XP
• Our own look at our logfiles

To be clear, those are all for internal measurement. And although some of that is state-of-the art stuff, no two really match up and sometimes there are even conflicts in reports within the same provider's system.

So then there's third-party measurement, which I'm not crazy about, but we feel that we have to use for two reasons:

1. Knowing what we know about the variances in all the metrics we've tried, we don't really trust the numbers that other publishers put out. Again, they may not be doing anything nefarious, but unless we know they are using Omniture configured exactly as we are, we've no faith it's apples.

2. Many folks don't publish the data at all. And it's critical that we be able to show something that helps advertisers understand that this upstart website has enough traffic to be worth their attention. Now our sales pitch has very little to do with total pageviews and who is the biggest -- we don't claim or seek to be the biggest site in town. In fact, with the Daily You, we strive to put many ads in front of as few people as possible, so that the advertiser doesn't waste money on a lot of untargeted impressions.

So then there's the question of what 3rd-party data we should use. Industry standard is Scarborough. But unfortunately, they're still on an old-media timeframe built around TV and radio that means that even though we launched in 2006, the earliest we'll show up in their survey is Spring 2009.

That leaves Alexa, which is panel-based and notoriously inaccurate, particularly on smaller (read: local) sites. Comscore is better, but is a closed and too-expensive alternative. Quantcast and Compete are hybrids. We err towards Quantcast because it comes closest to matching our internal metrics. It still undermeasures, but even though its stats for unique visitors are lower than Omniture's, we use the Quantcast number so we can make comparisons.

The panel vs. cookies problem is well documented and the general messy state of web metrics is too. Here's some good starting points, several of which are in conflict with each other:

http://www.digitalministry.com.au/com... http://www.pbs.org/mediashift/2007/08... http://www.naa.org/Resources/Articles... http://www.clickz.com/showPage.html?p... http://www.buzzmachine.com/2007/07/11... http://blogs.mediapost.com/spin/?p=1023 http://venturebeat.com/2007/10/22/fac... http://www.editorsweblog.org/multimed...

"can a third party site be trusted to tell you how many hits you get?"

A. What's a hit? We don't talk about hits, nor do most web publishers, partly because no one knows how to define a hit.

B. No.

"what i am asking is when this third party site says you get 280K visitors each month, do you trust that number as accurate?"

More or less. Our numbers tend to range 10-20k above theirs on uniques, which is close enough for me, given the alternatives.

you list over 1.8 million monthly page views on jan 1 ….. today quantcast says you have under 642K marking a SHARP DECLINE of over 200%!!!!!

There are a couple pieces to this answer. First of all, like all other web services, Quantcast is continually evolving. Back in January, they didn't measure pageviews. And, we don't generally make pageview comparisons between us and other media. And, our Omniture install was not complete at the time.

So we used a number we got from Google Analytics and our logfiles, doing the best we could to weed out bots, spiders, and the like (something that very few publishers take the trouble to do).

Now, our sourcing on the page should be clearer as to what is from where -- that's our standard, not general thinking. Many of our competitors don't provide any sourcing on the data in their sales materials. We do wherever possible, and admittedly should have done a better job here.

So no-- I don't think Quantcast is measuring pageviews accurately. That said, we've now got Omniture dialed in better and will publish a little more conservative number on pageviews in our update.

But it should also be noted that what our advertisers are buying is not our total audience, but the audience they, specifically, are being delivered. We give them access to the raw reports when they want it, so they can check us. But the real measure of delivery is results-- and we're building a nice group of testimonials that we'll also be adding to the media kit as part of the update.

"leave up the jan 1 numbers . . . it is sneaky sneaky sneaky"

You'll find very few publishers who update their data as frequently as we are-- prior to mid-year. The DMN, for instance, is selling using data dating back to 2005: http://adsource.dallasnews.com/portal...

And finally, as to:

"almost 2 days. no reply?"

Although Brett mentioned I was away, don't expect that we're generally going to fall over ourselves to answer you every time you sling a hand grenade-- We spend a great deal of effort to make sure that we're being responsive to our regular users. But you've joined very recently and contributed nothing but very pointed attacks -- from both of the accounts registered from your IP address. You haven't favorited anything or participated in the site in any other way. Without some idea that you're interested in doing more than picking at us, or perhaps some transparency about your identity and agenda, we can't neglect our active user community and our still-fledgling business to answer your every criticism. There's not another for-profit media outlet in town that would give you the opportunity to post this kind of discussion, unedited, nor one that would answer you so publicly and candidly. I felt it was important to answer this thread because of the false impressions you were propagating, but don't be surprised to be ignored in the future, at least until it looks like you're something more than an partially-informed sock puppet.

2 months ago ( Link to this comment | Suggest removal )