I’ve seen a lot of affiliates sink rather than swim on large network media buys. Many of them feel that this is the ‘holy grail’ to media buying, but this is far from the case. Before you can even consider approaching a large advertising network for your media buy (such as Tribal Fusion, Value Click, Yahoo, Burst Media, etc.) you need to make sure you’re having success on smaller networks (self serve) or similar websites. Just because you receive a huge influx of volume doesn’t mean you’re going to magically become profitable. Even when you port your ‘profitable’ campaign over to a large advertising network you need to be ready to have patience and optimize out the non-converting sites.
Here are some important items to remember before you even approach a large advertising network:
* Your campaign is already successful on a small-scale buy
– You’ve been making at least 100% ROI for two weeks on smaller advertising network/direct site buy. Don’t even consider a larger advertising network unless this has been met.
* Determine if the advertising network allows the use of a third party ad server
– With a large scale media buy, optimization is very important. Before you begin your buy you need to determine if the advertising network allows the use of a third-party ad server (such as Site S4cout or Zedo). Most of them do, but require you to sign additional contracts. By using a third-party ad server you can quickly eliminate non-converting creatives and scale accordingly. If the advertising network doesn’t allow third-party ad servers you need to make SURE you find out how soon they can make optimizations for you. Time is money with large-scale media buys.
* Determine a feasible out clause
– Most advertising networks carry a 24 – 48 hour out clause which means they have that amount of time to stop your campaign once you notify them. I personally like the 24 hour out clauses, as some of these greedy ad networks will hold you out all the way through the 48 hours (and you could be dumpy big money).
* Analyze your credit before applying for credit terms
– A lot of larger advertising networks are going to make you pre-pay your first $5,000 – $10,000 spend, however after that, they’ll normally put you on NET15 – NET30 invoicing terms. Before they grant you this, they’re going to require you to fill out a credit application. You want to make sure you list previous sources you’ve worked with and had a successful relationship with. It’s likely they will call all your references, so make sure the references are also aware of what you’re trying to accomplish.
* Test with one size creative and then branch out
– OK the 300×250 banner unit is most-likely going to yield the best click through rate for you. It always has for me and therefore it is going to be your most expensive unit regardless of the ad network. I like to start with this size and once it’s profitable, move into other sizes. Ad networks like to have you “bite off more than you can chew” just because it means more money to them. Get one size profitable then scale accordingly.
* Have plenty of backup offers
– If you’re doing a media buy for an affiliate offer you want to make sure you check with your affiliate manager on the cap room as well as alternative offers. Nothing is more deflating that getting an offer profitable and the advertiser pulling out the next day. Make sure you have at least 2 – 4 offers to test with and 2 – 4 backup offers. This will spread your risk and help you find a certain level of consistency.
* Pay special attention to the Insertion Order
– You’re going to have to sign an insertion order to begin delivery. I’ve already talked about the out clause but you also need to remember important items such as geotargeting, frequency capping, and day parting. Make sure you have the optin to disabled your campaign at certain hours of the day so you’re not running during non-profitable hours. This can normally be accomplished after a day or two running.
* Test a TON of creatives and landing pages
– For every 200,000 impressions delivered in a 24 hour time period I like to have two creatives. If my buy has 1,000,000 daily impressions I want to make sure I’m testing at least 5 creatives (if not more). You’re going to also want to test at least four different landing page variations. Make sure you closely examine the click through and conversion rates to the offer to determine the winning combination.
Buying media on a large advertising network can be VERY rewarding but at the same time can be VERY risky. You have to MAKE SURE you have your stuff perfected before taking things to this level. Feel free to ask questions below and I’ll get them answered as quickly as possible!
Ryan Gray is the founder and CEO of NameHero, one of the fastest growing independent web hosts in the United States. Ryan has been working online since 1998 and has over two-decades experience in Internet Entrepreneurship.
The largest networks in terms of audience reach –
Google – 156.1 million monthly unique – http://www.quantcast.com/google.com
Adblade – 147.8 million monthly uniques – http://www.quantcast.com/p-b8GPCpJxfqYm2
Meebo – 136.8 million monthly uniques – http://www.quantcast.com/p-93vmRJG_BQlqo
Vibrant media – 136.1 million monthly uniques – Comscore Oct 2011
Undertone – 125.2 million monthly unique – Comscore Oct 2011
Ask Network – 90 million monthly unique – Comscore Oct 2011
Kontera – 75.5 million monthly uniques – http://www.quantcast.com/kontera.com
Click Booth – 43 million monthly uniques – http://www.quantcast.com/clickbooth.com
AOL – 40.8 million monthly unique – http://www.quantcast.com/aol.com
Linkedin – 39.4 million monthly uniques – http://www.quantcast.com/linkedin.com
Pulse 360 – 39.6 million monthly uniques – http://www.quantcast.com/pulse360.com
Myspace – 14.8 million monthly uniques – http://www.quantcast.com/myspace.com
Really great layout of compartmentalizing (focus on one thing, then scale to others) and testing your profitable campaigns when you move traffic sources. Need to be reminded that testing and scaling never really ends. The only thing that we need to keep in mind is there is a point you can test too many permutations at once and lose too much money in testing. Just need to find the right balance.
Are you ready to scale that campaign to a million dollars a month?
Well, only wehan i have the bankroll to purchase huge amount of inventory…..
Till then, no 🙂
Some great points Ryan. I'd highly recommend in addition to the points you make that you request even delivery, so if you buy $30k for a month and the delivery rate or quality sucks and you want out, often they will speed up delivery in your 24-48 hour window at the end leaving with you little money left after the event. Even delivery guarantees that the most you will spend will be $2000.
Don't turn up at a gun fight with a sword.