Adwords - Initial Bidding Strategies for Affiliate Product Testing

Discussion in 'Adwords' started by gifmore, Jul 3, 2008.

1. gifmoreRegular Member

Joined:
Oct 12, 2007
Messages:
274
67
Hi BHW Folks & Adwords Experts,

Here's the case study: Let's say that I set up an Adwords PPC campaign that sends traffic to a domain of mine, which then routes the traffic to an affiliate offer that I am trying to test e.g. to see if I get a sale after X number of clicks. The direct linking and how the traffic gets routed to the affiliate salespage is not so relevant to the question, I don't think, so don't get too caught up with this process .

Anyways, if the affiliate commission is \$20 per sale, and let's say I asume a 1% (1/100) conversion rate such that out of 100 people who actually click my Adword ads and visit the salesletter, I am assuming that 1 of them will buy (hence the 1%). Based on this assumption and the commission pay out, my break-even bid per click (or max bid amount) will be \$20/100 = \$0.20.

The question is: What initial bidding strategies should I adopt, when i first set up a new ppc campaign, please?

So, say the max bid I should pay for, is \$0.20 in order to break even. If I bid this amount as the Maximum cost per click (CPC) and my ads show in positions 1 - 3, that's great . If the ads show in positions 4 to 6, I thought that's not too bad either But at a Max CPC of \$0.20, let's say the ads get shown only in positions 30 to 40, then that's not so good. and some might get shown in position 180 ....blaghhhhh.......

Given these possible scenarios, the real questions are:

1) Should I set my max CPC to be higher initally , than the break-even CPC, so that my ads are placed in better positions?

(i) If yes, which are the preferred positions to be in, i.e. positions 1 to 3, 4 to 6, or is this determined by how much higher a bid I am willing to set above the break-even CPC?

(ii) If no, and I should always set it at the breakeven CPC or lower, then, the concern is that, at ad position 20 or even worse positions (as decided by the breakeven CPC), it might take weeks before I get enough clicks (e.g. if I wanted 1,000 clicks to test this affiliate offer) to decide if the affiliate offer is fair and converts at 1% or better. How can I overcome this please?

2) If I do set a max CPC to be higher than the break-even CPC (e.g. \$0.50 instead of the break even bid of \$0.20), then how long should I leave it at the higher CPC level ? (e.g should I leave it till I get for example 100 clicks first or leave it for 48 hours, or what indicator tells me that it would be okay to reduce the max CPC). Just wanted to ask what criteria should be set for this.

3) Along the same lines, assuming the breakeven CPC is \$0.20, once I wanted to reduce my max CPC to an amount that is closer to, or equal to my breakeven CPC, should I
(a) just jump from e.g. a max CPC of \$0.50 back to \$0.20, or
(b) Should I do it gradually and in steps, e.g. from \$0.50 to \$0.40, then \$0.40 to \$0.30, and finally from \$0.30 to \$0.20?

Do my questions kinda make sense, folks ? Or should I simply set the max CPC at break-even point or lower, period ?

Thank you, all.

Cheerio

2. LanxRegular Member

Joined:
Jan 15, 2008
Messages:
350
102
shoot for the blue position, or the 1/2/3 spot in light blue color right about the organic search. It'll cost you a boatload more but statistically conversions are better in the 1/2/3 spot or anywhere in the blue, as opposed to the right sidebar.

3. multicolorguruJunior Member

Joined:
Jun 11, 2008
Messages:
176
24
Location:
Hell
In Search Network, always shoot for positions 1,2 and 3 first and a great CTR (which is also influenced by the position).

once you got the "reputation" meaning good CTR, you can slowly lower your bids while still keeping the 1,2,3 spots.

4. gifmoreRegular Member

Joined:
Oct 12, 2007
Messages:
274
67
Thanks, multicolorguru and Lanx.

I've gotten better CTRs when in better ad positions but still no conversions.

Cheerio

5. LanxRegular Member

Joined:
Jan 15, 2008
Messages:
350
102
that, or it's probably the quality of your keywords along w/ what you want your keywords to do. simple example
ipod
ipod review

are all extremely different and net different results for different types of campaigns, the keywords make the action themselves, you just have to make sure you want the proper action to be taken, look at it from a keyword level first, then see your add and landing page.

the keyword is the most powerful part of the equation do not neglect it with improper usage.

6. pdaguy9BANNEDBANNED

Joined:
Sep 5, 2007
Messages:
120
98
If you're testing, don't start with Adwords. Start with Yahoo - its cheaper and it will tell you what kws are converting. Once you get a profitable campaign going in YSM, then move it over to Google and shoot for the 1-2-3. Lowering your bids over time.

Make sure you have a content-rich site for QS on Google - Yahoo doesn't care so much.

• Thanks x 2
7. lotusJunior Member

Joined:
Jan 27, 2008
Messages:
108
25
If you are just doing initial market testing, don't bother with high bidding just aim to get positions 4-6. Bid the minimum that is required and watch how long it takes to get 100 clicks. This will tell you if there is a market and how much roughly it will cost.

Now bid high - if initial testing showed you got to positions 4-6 with 50 cents then bid \$2. Do this for about 2 days (or as much as you can stomach depending on the market) until your account has a nice high CTR (3% or above)

Make sure that your landing page has good copy - I know you said this was covered, but if it doesn't have the EXACT matches for your keywords and ad copy you'll pay for it. To get over this, either use a landing page script or use the fake url technique.

Fake url technique - Not sure if this still works but you can set up your ads to go to one url and the keywords individually to go to another. Basically you do a search in google for keywords in question and note the url of the number 1 organic search result. Use that url in your ads for both display and destination. Yes you will have to send those urls some free traffic. Wait a couple of days until adwords technicians have moved on and then go into the keyword editing of the adgroup and manually change ALL the keyword urls to your url. Hey presto great quality score.

Once you have a great quality score, lower your bids. If you are paying \$2 a click and you are not in position 1 think about other keywords. If you are paying \$1 but bidding \$2 and are in pos 1, lower back to 50 cents and monitor. Normally it's best to lower a couple of cents a day and monitor, but if you are breaking the bank with the \$2 clicks you can make a jump to \$1.5 then, \$1 then \$0.50. After you feel comfortable at around \$0.5 and provided you are still getting a good CTR you can start to lower a couple of cents every day and just keep monitoring until you see a difference in the clicks. Just remember its not what google says, its what you get away with. Sometimes you can put in 10 cents and google says your keywords are all inactive but low and behold you still get the same amount of clicks!

If you aren't converting clicks to sales adjust your ad copy. You are getting clicks right? Make the ads more specific or more attractive, don't use words like "learn" it discourages people. Use a call to action "click, download, now, etc" make sure you have some kind of urgency in the ad like "limited time only" or "today only" etc.
You might need to put up a "softener" landing page between the ad and the affiliate offer. Hope this helps!

• Thanks x 2
8. ronnyluaNewbie

Joined:
Nov 15, 2008
Messages:
3
0
That fake URL technique sounds great, but would the quality score revert if I change the display URL and the destination URL after that?

Or are you suggesting that we just change the destination URL of individual keywords and leave the ad unchanged?

Ronny

9. EntrepreneurBANNEDBANNED

Joined:
Oct 12, 2007
Messages:
439
379
I wouldn't worry about whether or not you start high or end high, the fact remains that two bigger factors in determining ad position and bid weight are the QS and relevancy of your LP.

Although saying that you're direct linking, so if i suspect the QS would be shit within a few weeks. The point is though if you bid \$0.20, you'll rarely pay that, obviously.

I'd personally bid different for each KW, within reason, if you're really working to a tight budget. If you have words like "Buy...", "Solution...", "....help" etc, then these will convert a shit load better than terms unrelated to a purchase, so would up those and drop others to compensate.

If you're getting a shocking ad position and you won't bid more, remove the word, it's just a waste of life. Also, I'd bid enough to get a good test and suffer a loss. Few Adwords tests come out smelling of roses across the board.

Keep the KW groups tight and relevant to each other, the ad and the LP. Once you've had x amount of clicks, go back through and remove words that don't meet your needs.

Get some Adwords reports done and check to see, what terms are hitting your broad matches, then add negatives (if you haven't already). Then based upon the broad matches, pull out terms that you've had a fair few times, and bosh them in their own Ad group as exact match and create an Ad focussed on those words.

It's a bit of a myth that one, sure it will improve you CTR, but if it's not cost effective that won't matter. Afterall, if you start low and come back up, you'd still be bidding less, albeit unnoticably, than if you started high with no CTR and came back down. It's just math: What's more 2% CTR of 100 then 5% CTR of 100 or 5% CTR then 2% CTR? They're identical. Technically speaking it'll make almost no difference at all because it's all even at the end of the day.

Also, test in 4-6; more wheat will be pulled from the chaf and it's less expensive. Given the choice, from what you say or imply about this product, at \$20 i wouldn't even debate putting it in 1-3 unless it converts sickeningly well. I'd probably also stick it in content exclusively first, because although conversions will differ and CTR will differ massively, you can probably build up a slight sinking fund to test on a larger scale in search.

If you've ran a test for a week or 2, your break even level will be more accurately known. Some words you'll get a huge CTR and no conversions, others you'll get a low CTR, but will convert at high rate. You need to test, it's all well and good using theoretical data, but the market will tell you the right figures, not a calculator and then you can adjust you bids and position accordingly.

Just test. You'll get real data and then you can have real answers. Just ensure you know what words are doing what, because some of them will sink you, while others will float you.

Yes, just change the destination for each individual keyword.

• Thanks x 1
10. ronnyluaNewbie

Joined:
Nov 15, 2008
Messages:
3
0
Sounds good. I'll give it a try.

11. bugschaserRegistered Member

Joined:
Jul 28, 2008
Messages:
88
29
Occupation:
IT Manager
Location:
Expat in Asia
Hi,

You might like to read the controversial but very interesting adwords report at: