Segmenting your inventory is important to get the best prices. Certainly for US-based advertisers, US-traffic is more valuable. We see lots of deals that are targeted as US-Desktop only.
Is it useful to break out other countries? Often, it’s not worth it! How could more money not be worth it? Well, it depends how much traffic you have and how much time it takes to service the deals.
Let’s say you have 10,000,000 monthly impressions. You earn an average CPM of $2.00 on these impressions. At the moment, you are not really segmenting geographically, however, you know that 10% of your traffic comes from the UK and it is also worth $2.00.
Suppose an advertiser comes along offering $2.25 CPM for UK traffic. Should you take the deal? At a glance, you might briefly think that $2.25 * 1,000,000 = $2250. Nice!
However, when you send those impressions to the new advertiser, you are no longer earning that $2.00 from the first advertiser. That is, you are making only $0.25 in new revenue. For 1,000,000 impressions, that’s $250. Not bad, but much less exciting than $2250!
Is $250 worth it? Ultimately that depends on your situation. If the deal is 100% fill, the setup is easy, the payment terms are good, and you trust the advertiser, it might be worth it. If any of these things are not true, this small CPM bump might cost you much more in the long run in terms of accounting, reporting, setup, debugging, discrepancies, and trafficking costs.
Every situation is different, but as a rough guideline, I would say do not consider geo-segmenting for less than 1,000,000 impressions.