• Jeff Lesser

    I think your study may be just as flawed as the self reporting. You analyzed the change history of people who signed up to have you manage their account for 30 days prior to signing up. Don’t you think they hired you because no one was managing the account?

  • Larry Kim

    Thanks Jeff! great point. as i said, it’s not scientific, which is why i so explicitly disclosed survey methodology. though, it could be the opposite. meaning the people who we sign up as customers *want* to to do better.
    they’re engaged and it’s a priority for them – you’d expect them to be a bit more active. there are other advertisers/agencies out there who don’t even care or have stuff on autopilot and are asleep at the wheel.

    In any case, there’s a huge gap between what is being reported vs. what is actually happening. keep in mind that i surveyed the same bunch of people for both questions. “how much work do you do” question was put to wordstream prospects in a webinar, and the customers which are a subset of prospects, so that cancels out at least some of the survey bias.

    Jeff – In your experience do you think it’s the opposite? that the average PPC account is getting attention every week? (not just your PPC account – i’m talking about the industry as a whole here).

  • Ronnie’s Mustache

    I agree, most accounts get neglected….and that is exactly what Google wants- IMO.

    Look at the top Ecomm advertisers. Many of them spent a lot LESS in 2012 than they did in 2011. When you pay attention, you often reduce costs.

    Amazon, Walmart, Target, B&H, US Auto, Barnes and Noble, Newegg, Crate y Barrel, 1-800- flowers.

    Those are $1MM+ per/month advertisers.

  • http://trung.tran.com.au/ Trung Tran

    If you setup the Adwords campaign correct from the start then half your work is done. As a minimum I would agree that negative matches and ad text optimization should be done weekly. The other activity that is not tracked is landing page changes which should also be done in conjunction with ad text changes so that the message is consistent.

    Over optimising can also be a bad thing. Continuously making daily changes to the same keyword, adgroup, campaign will make it difficult to analyze the performance.

  • Nebosh Course in UAE

    your collection awesome, your are giving do follow blog, make a no follow all the link otherwise your outbound link wiill be increased.

  • http://www.swydo.com/ jeroen maljers

    It’s an interesting study. Unfortunately the Adwords API is a but shaky on the change history front, otherwise we could define a Change KPI in Swydo . And do not forget that a lot of optimization work can also be done outside the Adwords interface (so without any registration in the change history):
    - Keyword research
    - Competitor research
    - Defining ad copy /usp’s/ tag lines
    - Configuring conversion scripts
    - Working in other interfaces (adwords editor) and uploading once a month etc.

  • Larry Kim

    the adwords api for change history works if you can figure out how to work around some of the (many) quirks. I’m guessing you’re hitting the *too many changes* exception – try splitting your timeframe in a recursive way (a bit like merge sort) until you find a small enough time frame for which changes can be returned, then add up all the pieces.

    Keep in mind that any changes made outside of adwords (3rd party tool like wordstream or adwords editor) will show up in change history when posting to the account.

    PS – i happen to think that change/activity is a great KPI for PPC. it’s only through thoughtful PPC optimizations over time that any gains can be made.

  • Larry Kim

    Hi Trung, thanks for this note. agree that over-optimizing can be a bad thing. However the data i have shows that people doing nothing far outweigh people who are over-optimizing (by 50:1)

  • http://www.swydo.com/ jeroen maljers

    Thanks Larry. I agree. That’s why we tried to build a KPI for account activity in the dasbhboard.

    We will try that splitting it up in chunks trick.

  • Ronnie’s Mustache

    They certainly do.

  • Dan Butcher

    Interesting article Larry. However, I feel there is another caveat that you need to add here for the readers. It’s all well and good making these points if the Adwords accounts are high-trafficked accounts – but if they are low budget accounts, or indeed accounts with very little search volume – optimizing on such a regular basis can be harmful to the account. For example, should you really reduce bids on a poor-performing keyword if it’s had just 10 clicks in a month? I would argue no. Statistical significance is key to any test, whether it’s in PPC or otherwise.

  • Larry Kim

    And that is why I graphed the activity by spend level. You can see there are lazy account managers across the entire spectrum of small to large advertisers.

  • http://www.fourstepstraining.com/customer-service-training/ps_exceptional-customer-service/ Tom Phillips

    As a small scale user of adwords I would try to review it at least every week. I find changing things too often actually makes campaigns worse. We get around 400 clicks total per month across a range of campaigns – some with maybe only 10 clicks per month. So we need to let the data build up before a decision is made to change it. When I have made changes on small amounts of data I have almost always regretted it later.

    The negative keywords is probably the area I put most regular focus on.