…and why the numbers still don’t add up.

So, I was going to drop this whole PJM subject, but then I saw a new comment in my previous post, “Where Did The Pajamas Media Money Go?”

What it reveals is that my estimates for how much they paid their bloggers actually overshot the mark. So that continues to bring up the question…where did all of PJM’s money go?

Without going into the specifics (it would violate what’s left of my contract), I can tell you that payment was made per 1000 impressions, and that a sliding scale was used. The most paid was $2. To reach that, one had to have over a million impressions in that quarter. Below 100K impressions, and your received nada.

What this means is that sites like IP and Malkin consumed an awful lot of the money. Good for them, bad for the rest of us.

Now, this sort of sliding scale makes perfect sense because it means that only the very top bloggers were getting a 50% cut of the lowest Above The Fold CPM rate ($4).

To put into perspective how difficult it is to get 1,000,000 impressions during a 3 month period, one would have to pull in 333,333 page views a month or 11,111 page views a day on average. In the 3 years this blog has existed, we’ve only gone above 300,000 page views once and that was at the very height of the political season last October.

So let’s take a look at somebody like Instapundit and see how much he was getting paid and how much Pajamas Media was making.

Sitemeter stats for Instapundit’s Q2 2008:
April: 7,979,546
May: 7,494,265
June: 6,986,270
TOTAL: 22,460,081

When calculating CPM payout, you have to take the page impressions and divide them by 1,000. That results in 22,460 for Reynold’s Q2. Multiply that by the CPM rate of $2, and you see that Glenn Reynolds had the potential to make $44,920 if the payments were guaranteed. And from all indications, that $2 CPM was a guarantee.

Over the course of a year, that means that Reynolds was making an average of $179,680. But if PJM was running their books correctly, they would have made $449,200 of revenue off of Instapundit’s traffic (given a $5.50 CPM, which is an average of their top rate $7, and their lowest rate $4) or $269,520 of profit after they paid him.

However…

If they only sold 50% of their ad capacity, then they were ultimately only able to make $2.25 CPM on Instapundit’s traffic or $202,140 in revenue. And once they pay Reynolds his guaranteed $2 CPM ($179,680), that leaves only $22,460 in profit.

But still…there should have been profit here. PJM never should have lost any money if their top earners only made $2 per 1,000 impressions.

Now, according to PJM’s own list of network bloggers, there are only 60 sites in the network. And probably only 20% of them hit that 1,000,000 impressions per quarter. Basically, a good rule of thumb is that you usually get 80% of your traffic from 20% of your posts/site/etc., so let’s just apply that reasoning here to PJM’s stated 29,000,000 cumulative impressions a month.

12 sites would generate 23,200,000 impressions every month at $2 CPM payout. Of course PJM paid out quarterly, so every quarter these bloggers would earn $139,200. Again, considering 50% capacity utilization for PJM across these sites at a $5.50 CPM should result in a net $2.25 CPM or $156,600 a quarter. Sure, it’s not a ton of money, but $120,000 profit every quarter? That’s certainly sustainable.

And let’s not forget that Roger Simon and Charles Johnson were most likely paying themselves ad revenue at the same time they were making profits from the cumulative network revenue.

So let’s look at the other 48 sites. They would have generated 20% of the network traffic, or about 5,800,000 impressions every month between them. And my guess is that the payout would be much lower than the top earners. I’d guess that PJM set a guaranteed CPM of $.50 or so since this is easily attainable with other ad modules out there.

And, as stated in the original comment above, some sites didn’t even get paid at all because they didn’t generate enough traffic. So let’s take the same 80/20 rule here for this group and say that PJM served up advertising 20% of the time and didn’t pay out a cent because those were on the 80% of blogs that didn’t get high enough traffic. So 80% of 5,800,000 is 4,640,000 of the impressions that PJM actually had to pay their bloggers for. That’s $6,960 a quarter. The profit on that would have been much more given the 50% cap util number of $2.25 CPM. Basically, $25,000 per quarter profit from the long tail blogs.

So let’s add it all up…

Assumptions:
- 29,000,000 impressions per month on all network blogs
- PJM ad network had 50% capacity utilization and any given time
- PJM ad network average CPM rate was $2.25 given 50% cap util
- $2 CPM guaranteed to blogs with over 1M impressions a quarter
- $.50 CPM guaranteed to blogs with under 1M ” ”
- 80% of impressions came from blogs that had over 1M ” ”
- 20% of impressions came from blogs that had under 1M ” ”
- 1,160,000 impressions were made that resulted in pure profit since traffic came from blogs with under 100,000 impressions per quarter.

PJM Projected Monthly Revenue: $65,250
PJM Projected $2 CPM Payout: $46,400
PJM Projected $.50 CPM Payout: $2,320

Which results in…

PJM Projected Yearly Payout To Bloggers: $584,640
PJM Projected Yearly Profit: $198,360

What does this tell us?

Well, the point of this post was to show everybody that PJM could have paid competitive CPM rates to their bloggers and still made close to $200,000 a year in profit. Obviously that’s not a gold mine, but I think it demonstrates that Roger Simon’s claim that the ad network lost money from day one is the result of extremely poor business decisions, and that all they had to do to make it work was tell their bloggers the truth and simply renegotiate CPM rates. That’s what every other ad network does when the numbers begin to go south.

But please, check my math and share your thoughts.