- The Search Agents - http://www.thesearchagents.com -

Some Google Bid Ideas are Google Bad Ideas

In the ‘Opportunities’ tab of the AdWords interface Google has begun offering a section called ‘Keyword Bid Ideas’ which provides two lists of bid change suggestions (called ‘Increase Traffic’ and ‘Decrease Cost’). Unfortunately, this feature is not currently available for all accounts and, even for those accounts where it is active, the ‘bid ideas’ are limited to keywords where Google’s Bid Simulator is active and which have not had their bid changed in the past 7 days. (Since I have yet to see Google’s Bid Simulator active for a keyword that has gotten fewer than 17 clicks in the past week and since The Search Agency typically updates bids for our clients on a daily basis, the number of keywords where Google is able to suggest a Bid Idea can be quite limited.)

In looking through their suggestions, though, some of these ‘Bid Ideas’ just seem to me to be bad ideas. Let me show a few examples.

First, in some cases, Google ‘Bid Ideas’ suggests both increasing a bid and decreasing it. Have a look at these three bid ideas from the ‘Decrease Cost’ list. To hide the identity of the client (a company that, appropriately enough, is involved in online privacy protection) I have deleted the ‘Campaign’ and ‘Ad Group’ columns and smudged out the names of the individual keywords.


For one word, Google is suggesting decreasing the bid from its current level of $32.00 to $24.00 in order to save $884.32 per week. However, on the ‘Increase Traffic’ list, Google suggests increasing the bid for this very same word from $32.00 to $35.20 to bring in 33 additional clicks per week at a cost of $1334.85.

Every search marketer understands that increasing a bid will typically bring in more traffic and decreasing will typically bring in less, but that’s true across an entire account. So why suggest both increasing the bid and decreasing the bid for this particular word? If Google’s ‘Keyword Bid Ideas’ are not specifically recommended courses of action, then what are they? (And this isn’t the only word I’ve seen this happen. I’ve noticed it for at least two other keywords in a different account, a firm that sells legal services, as well.)

In other cases, I have seen that Google’s ‘Bid Ideas’ sometimes suggests to push a bid down when it should clearly go up instead. For one keyword for the legal services firm, the bid was set to $5.00 and Google suggests dropping the bid to $4.55 to save $29.31 in potential cost. Clicking the blue hypertext of the Bid Idea pops up a list of possible bids and their associated estimated clicks, cost and impressions. In a previous post, How to Calculate Profit-Maximizing ROI [2], I showed how one can use this information, plus the recent historical Revenue-per-click and COGS-per-click, to determine the Value of the keyword (before advertising costs are considered), the Profit (after advertising costs are considered) and the profit-maximizing bid. This table is shown below for the keyword in question:


Since this word earns our client about $11.00 per click, a bid of $10.00 per click is estimated to yield the most profit. For the purpose of conducting an experiment I can’t describe in detail here, The Search Agency is currently holding the bid of this word to be exactly one-half of its profit-maximizing bid (This decreases our weekly profit by about $90, but we’ve decided that this short test is worth that investment). But Google is recommending decreasing the bid further to $4.55. This implies that Google is acting as if clicks for this keyword are only worth about $5 to this client, not $11. If Google knows that clicks average about $11 in value, then why is the Keyword Bid Ideas feature not recommending the profit-maximizing bid of $10.00? And, if Google does not have good knowledge of how much clicks on this keyword are worth this client, then why are they recommending bid changes at all?

Finally, I have also seen some Google ‘Bid Ideas’ suggest to push a bid up when it should be left alone. A different keyword in the legal services firm’s account earns about $14.30 per click (before ad costs are considered) and currently has a bid of $11.20, which even the numbers from ‘Bid Ideas’ suggest is the profit-maximizing bid. Yet, ‘Bid Ideas’ is suggesting to raise the bid for this keyword from $11.20 to $12.80.


If we wished to get about 55 more clicks but reduce our overall profit by about $100 for doing so, this bid idea would make sense. For that matter, we could trade any amount of profit for the specified extra number of clicks just by looking at the table of numbers the that Google’s Bid Simulator already gives us. So, again, we must ask ourselves: Why is Google suggesting this particular bid? That is, Why does Google presume to know that we are willing to trade profit for clicks and by how much?

The point is: Your business is your business, and nobody knows your business better than you. You know your value-per-click better than Google (or, at least, you should). Google’s Bid Simulator (the GBS) already gives ample information to calculate an optimal or near-optimal bid and, for that matter, The Search Agency was calculating optimal bids before the GBS came along and will continue to do so at least until API access becomes available for it.

While the bid change recommendations I’ve seen from Google’s ‘Keyword Bid Ideas’ feature seem reasonable in most cases (mostly because they tend to be modest in magnitude), there seem to be some unreasonable recommendations mixed in as well (as I’ve pointed out above). So far, the incremental value provided by Google’s Bid Ideas appears minor and the potential for lost profit seems high if the recommendations are accepted without scrutiny. Though Google will likely continue to develop the capabilities of this tool, at the present time it looks to me like some of Google’s ‘Bid Ideas’ are just bad ideas.

About Bradd Libby

Bradd easily becomes obsessed with small details. He can barely make it through a conversation without drawing a graph of something. He lives in the woods on the side of a hill in Norway. Bradd doesn't willingly wear shoes. He makes maps of places that don't exist. He picks up small shiny things he finds on the ground. He devolves into violent shouting matches with himself. He takes things apart to find out how they work but then can't figure out how to put them back together again. He also writes at Search Engine Land [9], Search Engine Journal [10] and his personal blog [11].