Crucial Mistakes from Media Buys

Hey guys, today I wanted to share some of my personal mistakes that cost me time and money in the beginning. I’d like you to avoid this at all cost.

Media buying is considered to be for big boys only. And it’s true to some extent. For larger volume, when we talk about direct/IO based buys, you need budget. Serious budget in some cases. But starting in media buys can be simple and easy. Sadly I had to learn it the hard way. So in this post I’ll give you some tips so you can save yourself some money and time.

Where to Start with Media Buys?

That’s where I screwed up really. Few years back in ’07 or ’08 I was hooked by Carlos and Lupe Garcia’s teachings and decided that’s what I should do. They talked about insane numbers and it seemed like I had to do it. So I hooked up with one large media buying network “Tribal Fusion” and made a $5,000 IO. Before that I didn’t know anything about it, but learned major tips from Carlos. 24hr out clause, even delivery, frequency cap and all that stuff. I guess I was lucky to know that, otherwise I’d lost much more money.

So I pushed some of the most popular products back then, some Amazon diet pills, and pretty much all diet stuff as it was extremely hot. And it was prior to Acai craze. So guess how I did?

$3K blown and 0 conversions. Nothing. I didn’t think it’s gonna be that bad, but it was. It’s not that the network was bad, it’s good. It’s just that when you know nothing in the space, you don’t go doing $5K buys. You start smaller. So where should you start?

Pulse360, Adsonar, AdBlade. Anything that has controllable interface where you can set your bids and budgets and not spend too much for a test. You can do Facebook and pretty much anything as it’s still technically called media buy. But if you want to do buys on high traffic sites, like major news sites, then these are the places to go first.

Never Trust Your Ad Reps

This one I didn’t learn. I talked to the ad reps and the guys sounded smart and cool. Suggested how to do it better, where to advertise, where other similar advertisers are advertising and so on. So I trusted him.

But when the buy started he didn’t do a shit. No management, no optimization. I didn’t know about adservers, so I used theirs. Wanted to save money. Big mistake. ALWAYS use your adserver. He sent me report once a week and I told him what to pause. No matter, the campaigns were still losing badly. So I asked what to do and the suggestion was to keep running.

Warning! Every fucking ad rep will tell you the same bullshit. You need to spend more, we need more data. NEVER EVER trust that shit. Always have a backbone and know what you’re doing. Tell them how much max you’re willing to spend for a test. If they don’t deliver, fuck ‘em. Cut and move on. There are too many advertising places to keep losing all your money on one of them.

So that’s another thing I’ve learned. Always have your adserver, be able to pause ads yourself. And if you see it doesn’t work, pause the whole campaign and exercise your out clause.

One a side note, it’s true that you need some data to make a decision, but it must be your call. If you only have $10,000 to spend and you do $5K IO on one campaign with 3 LP’s and 50 banners. Then calculate how much data you really need? Technically, each ad to each LP to each offer (if you’re rotating those) would take quite some money. Let’s count:

Your diet offer pays out $30.

You have 3 LPS and 50 banners. (Some suggest starting with 100 or 200 banners).

3×50=150.

150×30=$4500 for one campaign and one offer. If you rotate 2 offers, multiply that by 2.

That’s techinically.

In reality, it’s your call of course, but I would say if I don’t get conversions from 200 clicks to the offer, I’d stop the campaign. Something’s not working and you need to stop while you can. Now when bidding CPM it’s not that easy. You need to get good CTR. That’s why you need so many banners. You don’t run each banners for $30 spend. You just need to find the one that gives you the highest CTR. Which means your get the lowest CPC.

In large media buys CTR can range from 0.1% to 0.5% 0.2% + is good. 0.3% is very good.

So if your CPM is $3 on average. You manage to get banner CTR to 0.3%. You LP CTR is 30% and offer converts at 5%. You will need to spend around $670 to get those 200 clicks. The better those numbers above are, the less money you’ll need.

So prepare an Excel sheet before you do a buy, do your calculations and you’ll see how you’ll do. But don’t go doing big buys. You can do direct buys with smaller niche sites for $100 to $500 for a test. This almost always performs better and is less risky.

Ok, that’s pretty much it. These were my worst mistakes when doing media buys. I’ve done too many to mention here, but think carefully before blowing your money. Do your calculations and take as little risk as possible.

Good luck with your media buys!

Comments (2)
Add Comment
  • david suarez

    hey admin,

    I just rea dyour article, i am VERY happy i just read your article. I was going to do a 5k IO until i read your post.

    its funny that you mentioned carlos and lupe’s media traffic tactics. up until a few weeks ago, i tried desperately to find their videos of what they were teaching.

    As i researched about the industry further, I realized that the reason they were not offering their course any more was because it was complete outdated! with behavior targeting, DSP, RTB, etc.

    From what I read in your article, I can relate to – a lot. I am a twenty-something who gave up the cuckold 9-5 lifestyle and do things on my own. I got my degree in mechanical engineering, so im pretty good and comfortable with numbers. Thats why i have my attention towards media buys.

    If you can help me piece together pieces of the puzzle as far as where ad netowrks, ad exchanges, DSP, and RTB fall into place, you would be saving a lot of hair on my head.

    also, you mentioned in your article about “getting your own adserver” I def dont like the fact that if you use the adnetworks software for tracking what works and what doesent, they can see it and potentially leak out information or use it to their own advantage.

    If you could point me in the right direction as far as where to get aderservers or even code them ( i have a contact overseas that can code for me very cheap) that would also be huge.

    thanks again, and I look forward to hearing back from you!

    • admin

      Hi David,

      Glad I could be of little help 🙂

      Defintely, if you haven’t done serious media buying, then there’s no point going straight to ad networks.

      Here’s why:

      At best, ad network rep will talk you into buying ad placement in one of their categories, most likely they’ll throw in 5 or so for you just to start. If you have shitloads of money, that’s the way to do it. You gotta test a bunch of sites to see what works and what doesn’t. Your 5K or 10K IO won’t likely gonna give you any decent data. These guys have 1000’s of sites in each category. So it will blow your budget quickly and you’ll still need to test some more.

      Here’s what you do. Follow the money. First, I wouldn’t even start a campaign now without proper research. Figure out what works where. Go to major news sites CNN.com, newsweek.com, msnbc.com (properties of Pulse360). Watch the ads. Observe for a couple weeks. If an ad is there for 2 weeks on a high traffic site, it makes money. Copy the LP, copy the ad and create better ads if you can to split test. Don’t do RON, don’t do category targeting. Pick one site in the beginning and try to make that work. Or maybe 5 sites. If your CPA is $30 lets say (for skin or diet), then with 5 ads you need $150 max per placement to see if it works. Usually less. You can tell if it’s not working after spending like $70 bucks. Pick 5 sites like that and blow from $500 to $750 on each. That’s the best money you’ll ever invest.

      See, instead of blowing 5K on who knows what sites (ad networks don’t reveal exact sites, just number assigned to a site), you’re spending money on exact websites where other affiliates make money. This is the only way to enter the game David. It won’t neccessarily work. It’s easy to mistaken advertiser with affiliate, even if it’s affilite, they make get hihger CPA from a network, you may be scrubbed as a new pub… tons of situations why you may still not make money. But it’s the best way to do it.

      So Go to Adsonar or Pulse360 and start there. Pick only 1 site. If you’re starting out, noone will copy your campaigns, cause they won’t know if it works. They’ll only copy you if you’re running big volume ($1K/day) or so. In that time, you can hire some dudes to create custom tracking sytem for you with reliable cloaking. Don’t worry about that. Just test it out.