Quality In Search by Matt LeVeque is licensed under a Creative Commons Attribution-No Derivative Works 3.0 United States License.
Based on a work at www.semoe.com.
Aug
10
Stability, Capability & Goal Setting in Paid Search Advertising
August 10, 2008 |
ONE OF THE challenges I find most frustrating in Paid Search is the task of goal setting. Goal setting in an environment that has the capability of becoming unstable rather quickly - I am thinking a highly unorganized campaign structure here - and whose components of the process are out of ones control - things like search engine algorithms, after-click behavior, etc. leaves me sleepless in Philadelphia. To me these are quality related issues - trying to set goals in an unstable environment when the true capability of an account > campaign > ad group >keyword >ad creative is unknown could be identified as non-value added work or simply a waste of time. I am not in any way saying that goal setting is unnecessary because there are internal and external expectations to meet, I just think less focus on goals and more focus on process is a long term strategy that has greater benefit.
In order to justify my position on this topic I looked to advice handed down by W. Edwards Deming and one of his 14 Points of Management that I think applies to my case. In his book “Out of the Crisis” point 11b speaks to the elimination of numeric goals for people in management. The important parts paraphrased and responded to below.
Goals without method are burlesque. Examples: 1. Decrease cost by 10 percent next year; 2. Increase sales by 10 percent 3. Increase productivity 3 percent next year. A natural fluctuation in the right direction (usually plotted with inaccurate data) is interpreted as success. A fluctuation in in the opposite direction sends everyone scurrying for explanations and into bold forays whose only achievements are more frustration and more problems.
Wow does that sound familiar. I’d say mostly that the method of determining goals should be standard practise but will be modified from client to client. It’s when this method does not align with client goal setting methods things become fuzzy. The second part of this should be blatantly obvious especially when it comes to web analytics data - it is never accurate from application to application. Google Analytics will never align with with Omniture data and so forth. Trying to determine the ‘why’ in this is a waste of time - get over it, its the nature of the business. The other point here is ’scurrying for explanations’ also known as fire-drills. In a stable environment it’s easier to manage this part, but differences in cookie durations and other moving targets make looking at data in smaller time frames a waste. It is important to identify trends and potential red flags, but rushing to fix a perceived problem before it is a real problem can create bigger problems.
A man in the Postal Service wanted to increase productivity 3 percent next year. Enquiry about the plan for this task brought forth the usual answer: no plan - they were simply going to improve.
Planning is the first part of the PDCA continuous process improvement method. It is silly to even address this but it happens. I recently assessed an AdWords account for a small clothing retailer who had been spending $500 a month for the last 4 months without any analytics in place to messure success. When asked about how they were tying revenue to AdWords all they could tell me was that they were getting clicks and impressions… Clearly there was no plan or thought given to how to go about this critical step in managing paid search.
If you have a stable system, then there is no point in setting goals. You will get whatever the system will deliver. A goal beyond the capability will not be reached.
This I don’t entirely agree with since there is always is room for improvement. This statement would be more along the lines of a campaign or ad group in its current state. You’ll only generate x amount of conversions with a current set of keywords, ad creatives and landing pages. Expecting more without increasing budget, expanding keywords, implementing negatives or improving landing pages or user experience is a waste of time.
If you do not have stable system, then again there is no point in setting a goal. There is no way to know what the system will produce: it has no capability.
This is the most important point because I imagine there are very few paid search accounts that are 100% stable. Stability here means all ad groups are highly targeted, destination URLs are accurate and functioning, ad creatives are in the right position at the right time for the right customer who is ready to convert, all conversion tags are in place, reporting on performance takes as little effort as possible, etc. If anyone lives in this world please let me know.
I would not say that there is no capability though, it’s more like limited capability. Although there are times where ad groups or keywords just do not produce conversions regardless of targeting.
Focus on outcome is not an effective way to improve a process or an activity.
Again this is where more focus on the process and reaching a point where an account is more stable and has more capability is more important that focusing on false goals. It goes back to the whole Results vs. Process argument - focus on process and you will eventually meet or exceed desired results. Focusing on the number of keywords in an account or why an ad group had 2 fewer conversions from the day prior is a waste of time and takes away from the bigger picture at hand. Again it is silly to address these things but they happen far more often then they should.
This is by no means and easy task to overcome because its our unfortunate nature to obsess about outcome and results. There is a level of changing the way we think along with a level of education required to make the jump from results to process. This also takes a level of discipline and patience which does not integrate well with the ‘instant-results’ nature of paid search and online marketing. One recommendation would be to start small and gain the confidence of your internal and external customers. Another suggestion, especially in cases where things are so unorganized and out of control, would be to step back (or take however many steps back…) and reset the foundation of campaign structure and anayltics before moving forward with process that will eventually produce results.
Related Articles:
Paid Search Process or Results?
Uncertainty & the Age of Engage
How Stable is Your Paid Search Account?
If you enjoyed this post, make sure you subscribe to my RSS feed!
Comments
2 Comments so far
manufacturing quality…
In 1979, the British Standards Institution (BSI) developed the first commercial standard for quality systems that became known as BS 5750. That same year, BSI issued its first certificate to a small cement plant in England for compliance with BS 5750. …
[...] Stability, Capability & Goal Setting in Paid Search Advertising [...]