• http://www.charlesknightseo.com CSKnight

    Thanks for an excellent review, Alan. As a new SEO, I had to sort through some of these very same options.

    Looking at accounts much smaller than yours, I have a $105 setup fee($100 to set up the Google Adwords campaign plus Google’s $5 activation fee).

    I then have a bottom cap of $100 per month. I charge 10% of Google Adwords spend, which will
    become 15% when I pass the Google Adwords exam.

    I don’t have any maximum caps at this time.

  • http://www.topranksearch.com David

    I have a flat fee for Google PPC campaign set up at the moment – $2500 – if the client is happy I offer a drastically reduced monthly “retainer’ fee for tweaks and my professional eye.

    My reasons are thus if I set the campaign up at a low price I can get booted once it is done so a few days work and research makes me nothing….

    I do negotiate with Yahoo! and MSN PPC set up though.

  • http://www.linux-girl.com Asia

    I charge a percentage of spent dollars per month plus a standard monthly fee or a percentage of sales. I would prefer that the industry move into advertising agency standards, and agency rates are supported vs raising to standard fee charges.

  • http://www.feedthebot.com feedthebot

    As far as being transparent about costs, you won’t start a trend, but hopefully you will sustain the one that was started by SEOmoz when they discussed their pricing here.

    Actually, they didn’t discuss SEM specifically but I like that you and SEOMOZ and others are beginning to be less tight lipped about pricing.

  • http://pardonmyfrench.typepad.com Eric Frenchman


    This was very helpful. I often thought #4 was the best model and have been trying to move more clients towards it. Thanks


  • http://blog.thinkaboutsearch.com S Haar

    Interesting post Alan. My response to your assessment of a rev share is similar to that of your first article regarding branded keywords not being incremental – it is over simplified. We have been able to significantly improve the conversion rates for our partners on branded keywords (and non-branded) by taking ownership in a rev share model.

    The “click” in SEM is really only one part of a total package. What is the optimum positioning for end conversion? How are competitors challenging your client’s brand? How are you setting and addressing expectations pre and post click? What buy flow is optimal based on the iteration of the branded keyword?…and a host of other issues.

    With a rev share model, having skin in the game allows us to participate far more deeply in the entire process, thereby optimizing the program far beyond delivering a “click”. In a perfect world, all SEMs would do this. However, if you’re capped you are necessarily limiting resources. I do not believe this is good for either the client or the SEM.

    Any compensation model that distances the SEM from the ultimate goal eliminates the risk for the SEM. Sounds good, but it is not necessarily a panacea. As for who gets credit, there are ways to estimate the relative influence of each action. By understanding what an email subscriber (for instance) is worth based on a profile, email admin cost and average sale, even the future value of this can be assessed. Not to say that it is easy – it is not. But a rev share can be tiered and nuanced based on the client and the targeted actions, ultimately leading to the sale.
    about this last month in much more detail.

  • http://www.receptional.com Receptional

    So we charge a set up, then a percentage of spend (with a minimum cap) and then occassionally we taper the percentage at the top end rather than give it a ceiling.

    But I wish it were not so. It is not the right model as much as the model forced upon us by the agency world.

    The truth is that the work load is proportional to – mostly – the number of adgroups and creatives – rather than to a percentage of spend. If one creative gets 1,000 clicks per day then proportionally, the work was less than one that received 2 clicks per day… But my guess is that the 2 clicks per day cost much less on Google and much more for the SEM manager (in terms of time). The latter probably – statisticaly – has a better conversion rate too.

    Industry sectors are very different too. CPC proces in finance or porn are way higher than in travel or clothing, for example. So there is an argument to say that the percentage that the management company gets for porn or finance should be LESS than with travel or clothing.

    Maybe fixed fee with incentive overrides is a better model.

  • http://www.latitudegroup.com RichardGregory

    Great piece Alan but what about CPA based models?

    I also agree with Receptional’s comments that models can vary from client to client, or more likely sector to sector. For example, Travel campaigns are often more intenxive than finance campaigns.

    We have further complexity over here in Europe with the additionla element of search engine commissions/rebates being paid back to agencies spending suitably monthly amounts. How do you think that should be factored in?


  • http://sendtraffic.com sendtraffic

    There’s more then what’s listed regarding PPC pricing models. We use a flat CPC that’s based on our clients current PPC campaigns. This way, we start with what’s working and not have to increase the CPC by X percent. We put skin into the game that we can make a bigger, better performing campaign. If we can’t, we make no margin. Something to think about…

  • http://www.roiworks.com Porkchop

    Alan –

    Great post. One aspect of pricing models that does not seem to be discussed here, but is alluded to, is the size of the company as a factor in pricing.

    We have 2 totally different pricing models for businesses. If we are dealing with small businesses with under 20 employees (e.g. a lawyer, dentist, etc.) we charge a flat set-up fee (usually 399, 499, 599) and a flat monthly fee(also 399, 499, etc). A percentage of ad spend would not allow us to make enough off these little guys.

    If we have a larger client, or one in a hard to reach niche that needs a lot of custom keyword research, or one that may be hyper-competitive (e.g. mortgage loans), we usually charge a fairly sizable set-up fee to cover our up-front costs (10K -20K), handholding, etc. Thereafter, we either a) take a percentage of ad spend, capped at $X (depends on the client’s budget) or b) do a flat monthly fee somewhere in the range of 5-10K/mo.

    Hope this helps anyone out there. Please feel free to email me with any questions and/or to compare notes on our weird, nascent little industry.


    George Revutsky (Porkchop)