It's great that Jeremy is willing to share his numbers.
I would be shocked if his 2008 numbers were on par to $3.70. For a media presence as large as b5media, one would hope that both their traffic cycle and their eCPM would be growing, not remaining stagnant. While there's some economy of scale, large size adsense operations like Markus Frind's Plenty of Fish illustrate that a raw CPM of twice that (if not a magnitude more...) are completely reasonable, even across general interest blogs.
I'll throw the ball out there. I have one blog that I infrequently update (about once a week to once a month) that nets me $11.30 CPM in the health space. A small fraction of the traffic of any one of b5media's properties though!
I kill B5 Media. My blog's eCPM is over $100. If I ran B5 it would be making way more than $3.70 eCPM. That's a joke.
John: Of course your traffic is but a fraction. Hitting high CPMs on a couple of million pageviews is dead easy. Even on targetted verticals it's dead easy (we have verticals in the 30$+ range). But when you factor in high-traffic low-value areas like entertainment, it balances out to a much lower number.
To put this number in perspective, the last time I asked, Yahoo had 8$, AOL had 10$, Technorati had 8$ and Gawker had 6$.
And, yes, our CPMs have grown since then, but remain well in line with industry standard CPMs.
John: if you think 100$+ CPMs are consistently possible, you should apply for the CEO position at Yahoo. You could grow their revenue to a trillion dollars a year ;-)
Jeremy hit the nail on the head, it all comes down to economies of scale. The more pageviews you have, the lower the CPM will effectively be. Why? Because you will have unsold inventory if you keep your rates too high. It's better to sell all of the impressions than only a fraction. Of course you will still get high CPM's on the premium content pages, but the other pages will drag that eCPM down.
Glam Media, Tribal Fusion and other networks have an eCPM well below $3. Jeremy seems to be doing everything right.
I applaud the B5 folks for sharing their numbers. A couple of things stand out:
- Revenue/ writer is $ 6.6K. So, I can only conclude that this is a labour of love for most of the blog authors / editors.
Each blog (on average) generates $ 4.4K in revenue. Would they be better off focusing on a much smaller number of sites and driving critical mass to those sites?
Bottom line: the growth is there, and they're generating more revenue than many startups. Still, they have a long way to go.
Mark
Mark, good comments, however those numbers are over a year old :) (and in the original post, 2007 revenue is being compared to 2008 traffic which is even more flawed).
As a startup, we'll *always* have a long way to go, but our goal is to continue growing, continue evolving and continue pushing ourselves to never stagnate :)
Hey Jeremy,
If you are still open to being public, it would be cool to hear what the real numbers are. Don't blame you if you don't wish to share more. You've already gone beyond what most companies will share.
Mark
Hi Mark,
Probably not a courtesy to ask someone to publish their numbers nor to try and breakdown their revenue model in public. There will always be hidden intangibles that are not factored in.
It's fair to ask, but because we're VC-backed, we don't talk about current numbers. Normally we don't talk numbers at all, but made an exception for Profit, both because we felt we had a good chance of placing well and because we felt that so few folk share real details and as a result startups suffer.
And Ian, thanks for both comments :)
Ian,
I actually thought it would be better for Jeremy and B5 to have accurate numbers out there. If you see his comment above he is pointing out some serious innacuracies. Since these are public, I can only presume it's because he was OK with them being out there - in which case, better to have the right numbers out.
Mark
As a note, I have no issue with folk dissecting numbers or asking about numbers publicly. After all, I've done it often enough in the past. However the best I can do is point out the logical or factual issues behind those analyses.
Basically a lot of numbers are easily taken out of context, both because they lack context and because numbers from one time period are being compared to numbers from another.
We'll never disclose current numbers on anything but traffic, though, and even then we'll always keep it vague, so deep analysis of how we're doing will always have substantial assumptions built-in :)