If you've never attended a Search Engine Strategies conference, I highly recommend putting this on your "to-do" list. I first attended an SES conference in Dallas in 2000, and have been to many since in New York, Chicago, Toronto, Seattle and, last week, San Jose. The agenda is always fresh/current, while still covering the core areas of search marketing that remain unchanged.
One interesting session last week in San Jose was "Performance Pricing Models: What Every CMO Must Know." The session was moderated by Andy Atkins-Krüger, from Web Certain, and the speakers included Paul Wilson, chief revenue officer at iProspect, Andrew Beckman, president of Location3 Media, and Vivek Bhargava, managing director of Communicate 2.
This topic interested me because I often reflect on my past in this industry and well remember the days of selling search advertising for Lycos, at CPMs that were anywhere from $30 to $150, and then had some company called Overture come in and begin offering CPC pricing. The effect of performance-based pricing for ads was immediate and dramatic.
I've always kept an open mind about the possibility of performance-based pricing, but SEO has so many variables to consider that it can be incredibly difficult to pull together. The panelists seemed to agree.
Wilson said that 95 percent of their deals remain monthly SEO management fees and a percentage of spend for management of paid search initiatives. He said that 5 percent of their initiatives have "some type of incentive built in."
The audience included many marketers. Not surprisingly, the vast majority of these people were interested in performance-based pricing models.
With performance-based pricing models (those that don't include at least a base fee, with incentives) the risk goes to the agency. With organic search, the process can take time (especially if you want your efforts to be as "natural" as possible). So how long can the agency float this?
Also, Google could have a dramatic algorithm change (Google Caffeine?) that affects rankings/traffic. Is it fair that the agency has to spend a lot more time -- perhaps for little or no compensation -- because something outside their sphere of influence changed?
Even knowing these challenges, I've wanted to do this for some clients in the past. I just had a difficult time getting a model in place that absolutely works on many levels.
A Performance-Based SEO Effort
First, you need to have very solid metrics. These metrics begin with at least 12 months of solid analytics data.
Many times, I've partnered with a company and, upon closer inspection of their analytics installations, noticed that they have corrupted or inaccurate data. They either didn't break out their paid search traffic from their organic search traffic, or they didn't have things like Google Product Search broken out, or they didn't get tracking code on every page of their Web site. Because this data hadn't been correct for at least the past 12 months, how can any discussions on a performance-based SEO continue?
Assuming that the analytics is installed properly (and both parties agree to the third-party source for this data), you need participation, collaboration, and cooperation from your clients to see SEO success.
If you consider a performance-based initiative, how many of the deliverables do you directly control? If the Web site needs content, who is going to write it? Should the agency carry the load, especially if they don't have a monthly retainer to offset the costs? If the client is carrying this responsibility, how do you ensure they do what's necessary for success on their end, and in a timely manner?
Metrics of Success
In order for any SEO effort to succeed, you must first define success. Is it a certain lift in organic search traffic over a given baseline? Is it a number of leads? Is it sales?
This is especially important if you're considering a performance-based initiative. If it is an increase in organic search traffic over a given baseline, I think we can all agree that some keywords are more valuable than others.
That's why, in AdWords, you have a different CPC for different keywords. Some keywords might be worth $0.10 per click, while others might be worth $35.
If we assume that a client has analytics installed perfectly, and tracking data perfectly for the past 12 months, and that they have the dedicated resources for copywriting, implementation of recommendations, and any other considerations that the SEO firm deems necessary, are we going to be able to get credit for "everything" that the SEO programs brings, in terms of value?
I asked one panelist why they completely discount the traffic coming to a Web site from "branded" terms. "Branded" in this case means that someone visits the Web site after searching something like "company name" or "company name/service."
Now, I "mostly" don't measure success against branded search traffic. But I don't ignore it either. Branded search often increases with an SEO effort. I think it's safe to assume that the more presence you have in the search engines, the more people are exposed to your brand, and the more often they will search for your brand.
Branding isn't a bad thing. Those people who search branded keywords can convert pretty well, so I don't think it's bad traffic. What percentage of that should an SEO take responsibility for? Great question. I don't know the answer. I don't know that anyone has that answer.
This is where it gets very difficult to entertain performance-based pricing for SEO. This must be something that is good for the agency (has upside for them) and good for the client. To get something that both parties can agree to, you have to look at a multitude of factors: how success will be measured, what analytics products will be the "trusted source" of defining success, who will do some of the leg work, how responsive both parties must be to pushing the initiative forward, and many more questions.
Always select a firm based upon qualifications, not on whether they would do a performance-based initiative. Beware of "giving an agency a try" because "it won't cost us anything."
In some cases, SEO firms have done more harm than good. I don't say this to scare you -- just to make sure you cover your bases and do your homework.
Performance-based pricing might have a chance of working for paid search (especially the hybrid model mentioned earlier), I just wonder how a truly fair performance-based initiative can happen for SEO.
If you have an experience that you'd like to share, please post a comment below!
Twitter Canada MD Kirstine Stewart to Keynote Toronto
ClickZ Live Toronto (May 14-16) is a new event addressing the rapidly changing landscape that digital marketers face. The agenda focuses on customer engagement and attaining maximum ROI through online marketing efforts across paid, owned & earned media. Register now and save!