Welcome to WebmasterWorld Guest from 3.209.80.87

Forum Moderators: buckworks & eWhisper & skibum

Message Too Old, No Replies

Start by bidding High or Low?

How do you start off a new campaign?

     
3:49 am on Jan 20, 2005 (gmt 0)

Junior Member

10+ Year Member

joined:Nov 21, 2004
posts:62
votes: 0


When starting a new campaign, do you start off by bidding high, and then slowly lowering your bids to see what you can get away with? (Of course starting at a price you think will still have positive ROI).

Or do you start at the minimum CPC and then try to gain more traffic little by little by upping your bids and not blowing your budget.

I've been doing the second method mostly, because I'm still relatively new and want to cut down on "too expensive" lessons. But I can see a lot of merit in doing it the first way, especially now that I've had some success and my "gut-feeling" is a bit more in tune with what works and what doesn't.

Thoughts?

-V

9:33 am on Jan 20, 2005 (gmt 0)

New User

10+ Year Member

joined:July 15, 2003
posts:40
votes: 0


Hi

This is a very good question and one that I'm also wondering...

If Max CPC x CTR = rank number... then new campaigns with a virtually 0 CTR are going to be ranked low down.

With a low ranking, boosting the CTR is hard because you ad is featured in low positions and get less clicks!

So maybe the answer is to payout top money initially, get a good CTR and then reduce the MAX CPC once the CTR has climbed?

Anyone had experience of this that can tell us the best way to go about this?

Thanks
Dave

9:46 am on Jan 20, 2005 (gmt 0)

Preferred Member

10+ Year Member

joined:Oct 23, 2003
posts:353
votes: 0


Give them a boost, it works!
10:06 am on Jan 20, 2005 (gmt 0)

New User

10+ Year Member

joined:July 15, 2003
posts:40
votes: 0


Thanks.. Now if i stick my MAX CPC very high what price am I actually going to pay?

In the google estimates its saying that on a certain phrase the "Average Cost-Per-Click" is very low 0.04p...

Does this mean if i put my MAX CPC up to a VERY high amount i will ever pay the high amount? Surely with google estimating 0.04p its never going to get as high as 1 if i happen to set my MAX to 1?

How does google actually work out what the click is going to cost?

10:45 am on Jan 20, 2005 (gmt 0)

Preferred Member

10+ Year Member

joined:Oct 23, 2003
posts:353
votes: 0


Well i think that you will have to pay more than what you might normally, but that is just to get position. In this game CTR is seriously important, so be prepared to make a loss initially, and claw back that loss later.

I can't really tell you how much u will pay, suck it and see on a couple of words.

10:57 am on Jan 20, 2005 (gmt 0)

Preferred Member

10+ Year Member

joined:Oct 23, 2003
posts:353
votes: 0


CPC is worked out by taking your bid and looking at the rank number of the nearest competitor, something like that anyway. If you bid 20 dollars on a non competitive kw there is little chence of you paying that, and ieven if you needed to, it wouldn't be for long.

Actually there are loads of threads here about working out the actual CPC. And those guys are much better at explaining than me! My only advice is get hooked on your CTR.

Good luck

11:38 am on Jan 20, 2005 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:May 2, 2002
posts:928
votes: 0


About to launch a campaign for 1 small-ish client targeting 30K words. We're starting at 5c each to check the volume of impressions but using a very high daily budget so as our ads show.

Once we have the impression volume data, then we're in a far better position to rebuild the campaign targeting the KWs that work.

12:48 pm on Jan 20, 2005 (gmt 0)

Preferred Member

10+ Year Member

joined:Oct 23, 2003
posts:353
votes: 0


that's a good idea for some. But the keywords we go for dont get many impressions at 5 cents....5 cents and they wouldn't even get 1 impression - ever!
3:14 pm on Jan 20, 2005 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Oct 8, 2002
posts:657
votes: 0


Unless the client needs to get to market fast, I usually start with a bid of about 75% of what I estimate the average bid will be. This is enough to achive delivery on the cheaper keywords. I then individually bid these keywords to the appropriate amount, then raise the default bid incrementally until I achieve delivery on the most expensive keywords.
3:24 pm on Jan 20, 2005 (gmt 0)

Moderator This Forum

WebmasterWorld Administrator ewhisper is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Feb 13, 2003
posts:4744
votes: 8


There are a few different approaches, which depend on the site's goals.

1. Traffic. For this, the above approach of a low bid (not necessarily $0.05 - you don't want to be buried so far behind you never receive an impression) to find where volume and quality lies is a good approach. Measuring time on site, time on page, ad (on site - not G) CTRs, etc help determine which KWs are being successful.

2. eCommerce (or Goal oriented marketing. This might not always be eCommerce, it could be newsletter signups, free downloads, etc - but there are specific user actions that are measured and can have $$ amounts assigned to those actions).

There is a need to to play with who's around for these sites. For this, we look at conversion rates, profit margins, volume, and then make a competitive bid (not necessarily the top premium position out of the gate, just something that will produce a break even or positive ROI). Once we have some data to work with, we can start adjusting bids based on conversions, ROI, LTV, etc.

3. Branding. Need to promote your site in the top 5 for a decent branding effect (top 5 if 2 premiums, top 3 if 0 premiums -- that is for a 'real company' who has some sort of name already, if you're a small company who wants some branding efforts, a lower placement can work as well as image ads). Often these guys have a bit more money to play with and aren't looking for straight ROI (sometimes they are, in that case this is used with a mix of the eCommerce above).

We make competitive top 5 bids, and then measure CTR, average times on site, average times on pages, etc to determine which KWs/Themes will help the branding effort, and which KWs are so expensive - it's not worth their time as they aren't looking for ROI, but total exposure at a reasonable price.

Every site is unique in it's approach to marketing, conversions, defining & measuring success, that it's difficult to lump all the variations into a few categories. The above are the top 3 I run into, but there are plenty of other sites with other goals that need to be taken into account as well.

The questions to start with:
1. How do you measure success?
2. What do you consider a successful online campaign?

From there, you can use your knowledge to help them realize if their goals are realistic, if they're missing a goal they should be looking for, etc.

6:36 am on Feb 2, 2005 (gmt 0)

New User

10+ Year Member

joined:Jan 17, 2005
posts:4
votes: 0


I agree... to a point.

I strongly believe that 'branding' for a small company is not an option.

spend your money instead on creating MASSIVE value for the prospect. offer them something of value in exchange for their contact details (the more the better, but email & name as a minimum)

Then follow up. again & again & again.
prove you can provide tons of value
prove you can solve their problems.

unless you're Coke or Unilever or Nike branding ain't gonna work.

Permission Marketing is king. get their permission!

Wow them, knock their socks off, make you're offer insanely remarkable.
Otherwise don't bother...

just my thoughts :)

Mikey R