Watch Out for Runaway Bids

By , President at Hanapin Marketing

132 SHARES

Two weeks ago I had an alarming discovery in an account. Based on the geo-targets, day-parting and device based bid multipliers I had turned 300,000 keywords in to 7.4 million UNIQUE bids. The reason, as many in the PPC world know by now, is the multiplied affect of bid multipliers.

The math is pretty straight forward. You take the number of bid multipliers for location multiply that by the number of mulipliers by schedule (day-part) and then you multiply the result from that by the devices (if you are using mobile multipliers this number will always be 2, if you are not then it will just be 1).

So that’s: (location multiplier) x (schedule multiplier) x (devices) = total bid multiplier

 

Here’s a simple illustration of an account with two locations, two ad schedules and bidding down for mobile:

 

AdWords Bid Multipliers

 

This chart represents one keyword with a starting bid of $1.50. You can see that it turns into 8 unique bids (on the far right) with only one being the original bid. I see two ways to combat this:

 

1. Keep locations in separate campaigns still if you are going to use the multipliers.

This is the opposite of Google’s advice but location multipliers are likely to be the most varied and thus create the most number of unique, hard to control actual bids.

 

2. Use a chart like above to see what your total multipliers would be and for any with a range greater than 30%, put them in a new campaign.

To do this, add the columns I have listed above in the chart (Original Bid, Location, Location Multiplier, etc.) to a spreadsheet. Then open the settings tab in one individual campaign and input your multipliers. Currently this is a manual process but I am hopeful that Google will allow a download feature soon.

A quick note, although in AdWords a bid of 100% is actually bidding up by 100% (so doubling the bid) for this sheet you want to use 100% for setting it at 100% of the original bid and then use number like 1.10 for a 10% increase and .90 for a 10% decrease.

Once you have the data in there do the formula (Location multiplier cell) x (schedule multiplier cell) x (device multipliers cell) in the TTL Multiplier column That will give you the Total Multiplier percentage, or the actual change in the original bid. Then in the final column multiple that TTL Multiplier by the far left cell, which is the original bid. That will give you your new bid.

Once you have that you’ll want to find the range of your total multiplier. For that you take the highest percent increase and subtract the largest percent decrease. Using the chart above, the highest increase is 110% and the lowest is 76%. That equals a range of 34. So in this case I would want to consider taking the highest or lowest of these and put it in it’s own campaign.

The reason I do this is that it is hard to make sure you show in the position you want, at a CPC you want when the bid for one keyword could be 30% higher or lower depending on your multipliers. There is just too much variance there for my liking and control freak nature.

 

So be careful! Using the new bid multiplier in Google AdWords is a powerful tool but not considering all the variations of an original bid could lead to cutting out volume or bidding substantially higher than is profitable to do so.

Get more weekly links with our Fast Five newsletter! Five Fast Links in Your Email Every Friday.

Also send me a daily RSS digest

Set a Realistic Budget for Your PPC Account

Twitter Facebook LinkedIn Google+ Email Print More
  • http://twitter.com/sammmer Sam Mazaheri

    I propose “taptop” as an alternative to “desklet”

  • http://www.facebook.com/erika.p.mccarthy Erika Parker McCarthy

    Any suggestions for scaling this for larger accounts? I like the control you are creating here, but we are talking about a lot of work and campaigns to manage if you are breaking out everything with a 30 pt. variance even 1K keywords.

    • http://twitter.com/JeffAllenUT Jeff Allen

      The account I used in the example had about 40k keywords and separating those with 30%+ variance would only add 2-3 campaigns. That’s because most of the ones on the high side (so +% of variance) could go in a campaign for higher than normal keywords and keywords that are on the lower end (-% multiplier) of the multiplier go in a campaign for low multipliers.