There are PhD infused automated systems, handmade data models, and the old-fashioned "gut feel" approach. Is any one method better than another for solving the data-driven puzzle of paid search?
Every bid management system has its advantages and disadvantages. The key to paid search success is being smarter and quicker with bid changes, especially considering your marketing spend is directly tied to the set values per click.
However, since the introduction of quality score, bid management is no longer the difference between a 1 ROI and a 15 ROI. Back when Overture was king and advertisers paid just $0.01 more than the next highest bidder, you needed to mindfully watch bid management. You didn’t want to get squeezed by someone bidding a penny lower than you. This made watching your bids 24/7 a key strategy.
In today’s world of relevancy and quality score, many more factors impact your CPC. There’s CTR, landing page quality, relevance of the keyword to the ads in its ad group, and many more. So when you change bids, you’re only moving one lever of many to impact performance.
This isn’t to say that bid management isn’t important. We spend a lot of well spent hours building, designing, maintain custom algorithms to meet our clients bid management needs according to various goals and metrics, and this is time well spent. The idea here is that bid management isn’t a silver bullet.
Bid management will help make fine tweaks that aid in determining optimal position, given set goals for volume and return. Just like free market economics the marketplace will determine the bid price. While you can choose if you want to play on the high, middle, or low of that market price, you’re going to pay that price within a range.
Take Apple stock as an example. This week, it was trading at $235, just off the 52 week high of $237. If you want to buy Apple stock you can put in an offer price of $200, but the going rate ask and bid price is only within a dollar either way of the market price. Meaning, even if you get a deal of the stock from someone who is bearish, you’re only getting a deal by 0.43 percent. That can be exciting, but unless you’re trading millions, you aren’t retiring on that savings.
Let’s look at an example using three scenarios, and assume the goal is a 2.5 ROI.
In Scenario 1 with pre-bid optimization, the return was 2.2. There are many ways to approach this problem, but let’s focus on bids vs. keyword selection to see what the quickest way to the goal ROI is.
In Scenario 2, all bids are reduced by 10 percent (assuming the maximum CPC changes and all other elements are constant). The impact to ROI is an increase of 11 percent, to raise the overall ROI to 2.42.
The bid change made progress, but in Scenario 3 we remove the two lowest performing keywords (B and E), and the impact is 40 percent, to a 3.05 ROI.
This is where a great healthy debate should start to occur. This example assumes that the only goal is to meet, or exceed the ROI target. It says nothing about revenue goals, or a marketing budget.
Often these items are at odds with one another. These tradeoffs are critical to understand. However, when looking at the quickest way to a better ROI, it’s not in lowering bids by X percent each day until you get there — it’s by removing the poorest performing keywords.
You should be looking at many other things that you can read about across this site to fix those keywords, rather than removing them. Was the right landing page being used? What ad copy and match types were at play?
These questions help create the great data-driven puzzle that is paid search, and is what makes talking about this stuff exciting.
Join us Tuesday, April 6, 2010 at 8 a.m. EDT (1 p.m. GMT) for a free Webcast, “Reaching Customers Wherever They’re Browsing,” which will give you a better understanding of how contextual advertising works. And using the right management tools you’ll learn how to plan, execute, control and monitor campaigns for maximum ROI.