Forum Moderators: open
What do you need to know to dertermine your bid?
[assumes site sells stuff]
[assumes this is not a branding exercise]
Your site's conversion rate comes to mind.
Oh man! Can I point you to about 200 sites that are caught up in a frenzy of "acquiring traffic" and haven't a CLUE what they're going to do with it? One site comes to mind in particular. For starters, they are bidding .17 on a regional geographic (resort area) term when the #2 bid is something like .03. Once you get to the site, it's a Links2-style directory hoping to get advertising from local businesses in the region. The ONLY advertising on site is a banner for an art gallery and guess who owns the site? Yep, the art gallery has now spun off a web directory business. If their intent is to get advertisers/sponsors, they are waaaaay late to the party. If they're attempting to convert to art sales, their decidely UNfocused traffic is likely to have a conversion rate of .000008. At their position, I figure they're blowing a minimum of $35/day -this will easily hit $200/day in the tourist season (unless the account goes dry before then).
>For starters, they are bidding .17 on a regional geographic (resort area) term when the #2 bid is something like .03
Got to get in there and LOWER your bids on a regular basis or you'll be left hanging like that guy/gal.
Just like dirt world direct response - Mail out x pieces, at y conversion rate = z sales at Q margin. Subtract the cost of campaign (printing, postage, creative, shipping, cost of goods sold, etc.)and see what you've made.
BTW in dirt world direct response, 1.5 - 2% conversion is considered very good. (50:1 looks-to-buys ratio)
Let's say your conversion rate is 3%
Your profit margin is $20
What is your MAX bid on the CORE kw?
[opinions may vary]
What if you have more than one product/service?
That holds good for traditional methods, the difference here is that the "x pieces" are at variable cost. IMHO opinion the PPC market needs to be approached in a more campaign like manner, akin to the traditional advertising world, where you have a fixed budget to spend in a timescale. The challenge is to get the maximum traffic for that cost, individual bids are just a constituent part of this, the whole story is the campaign cost.
>What is your MAX bid on the CORE kw?
Depends on your MIN.
Now what do you think would happen if you tweaked one variable?
Say that you spend some money to completely redsign the site.
Make it more profession, well written, user friendly, great FAQ etc.
If you could increase your conversion rate by just 1%......
I like to test two different click-through pages, always keeping the one unchanged that converts the best. Then I tweak the underperforming page to work towards improvement, but the performer is still out there performing.
On GoTo, this can mean changing the clickthrough URL very frequently to try to even out the sample. More expensive advertising methods -- such as banner ads on search engines -- allow for the automatic alternating of two different versions of the materials. I love that. I get to rotate two banners on the same keyword and also alternate the clickthrough page for each banner. I get a nice 2x2 results matrix that really helps zero in on what's working and what isn't.
Along these lines, in some campaigns I create a separate clickthrough page for each keyword bid on. By doing this I've discovered that conversion rates can vary greatly from one keyword to another. So when it's practical, I don't throw all the clicks into one pot and lose that highly useful statistic. Several times we've ended up dropping the "obvious" keyword bid, because it doesn't give traffic that converts nearly as well as another, less obvious, word.
RE: Max bid core KW at 3% and a $20 margin? Well that would depend on your "required or expected" profit. A max bid of .60 would yield a total cost of 19.80 (33 clicks at .60)to make the $20 sale. Slim pickins - but still a profit. Personally, I would max out the aggregate of all bids at about half that, or 30 cents, yielding a click cost of 8.90 for the same $20 margin, or a 11.10 profit. Other folks use a different formula showing cost versus value of a visitor. If it takes 33 visits (100:3 or roughly 3 percent conversion)to make a $20 sale, a visitor is valued at $20/33 or 60.6 cents. If your cost per click is an average of 30 cents, you "make" 30 cents per visitor in eventual sales - provided the conversion stays constant, the terms do not lose relevance by being too obscure etc cetera. I aim for 2:1 value to cost. If you can perfect a system that yields $2 cents for every $1 put in, lookout Bill G. As a gambler, I have this feeling that such a system will be violently skewed by terms becoming more competitive, and the cost rising. Spending more to acquire a visitor than they produce in net magin is silly. This cost can be time in SEO, PPC, CPM in banners, etc. IMHO, for now, PPC is the best value, and the most immediate ROI. For the long term - barring unforeseen algo shifts, SEO is better and cheaper. Neither may be enough to make the big guys profitable, but both are great for the smaller operators. We've had many large dotcom clients tell us that PPC is the most effective thing they've done.
NFCC - re: the campaign cost focus. It does look like a straight media buy. As for the cost per click variations, we use an aggregate cost, with a maximum, and no minimum ;). we use a formula that very accurately estimates traffic through each term according to position. For ease of operation, the descriptions are usually identical. We tend to do long term campaigns this way, as long as they are profitable. Campaign style focus is great for building brand, mailing lists, and temporary boosts in sales. If we isolate really low cost clicks, we ride them out until they rise above the net value of a visitor. YMMV
Great post!
[wonders about - the descriptions are usually identical]
