• Jason Pomerantz

    Hi George, great post, I would just add one point.

    It does help to have a budget for testing. Since most tests will fail, it’s good to know in advance how much you can devote to playing with keywords and message, with no expectation of positive ROI.

    -Jason Pomerantz
    VP Digital Media
    R. A. Rapaport Publishing, Inc.

  • http://www.rimmkaufman.com/ George Michie

    Jason, thanks for your comment. I agree 100%, and should have mentioned that. Gets to the guardrail metaphor: the engine budgeting mechanism can absolutely save you from disaster for new initiatives; you just shouldn’t continue to hit them as a matter of standing practice.

  • Pat Grady

    Once trust and performance are established, we seek “Soft” budgets, ones that can be reasonably (~30%) exceeded, if ROAS / CPA goals are at or above goal. Fixed budgets can lead to missing out on big opportunities (seasonality, a competitor’s site offline, traffic spikes from many sources, new experiments, etc).

  • http://www.rimmkaufman.com/ George Michie

    Smart approach, Pat.

  • http://www.andykuiper.com/ Andy Kuiper – SEO Analyst

    …we’ll have to run it by (insert big shot’s name here) and see if we can get an ok to increase the budget —> forget about making sense LOL. Great common sense topic here George, thanks for posting this :-)