Performance-based Pricing for SEO won’t work
Ian Lurie Aug 10 2009
People love to come to me and say “I’ll pay you $X if you get me the #1 spot for ‘nude chicks'”. Yeah, great. I’ll jump right on that.
You can’t price SEO, or any other internet marketing, or any other marketing, for that matter, based purely on performance. Oh, you can try. It sure simplifies things. Instead of having to think about stuff like brand, long tail keywords and team interaction, you can use a simple “sucks/doesn’t suck” model.
It simplifies things. And it’s a crock. Here’s why performance-based pricing doesn’t work for SEO, or most other marketing:
You won’t give me 100% control
I won’t even consider performance-based SEO without 100% control of the web site. That’s 100%. Of everything. The shopping cart, the content management system, hosting, content, design and layout.
Think about it: SEO depends on onsite optimization, link building, trust and other factors.
I can’t optimize your site if your development team refuses to make 1/2 my changes.
I can’t build links if you won’t let me post quality content.
And I can’t build your TrustRank if your server is supporting DDOS attacks against Twitter. And I can’t deal with that if I can’t get into your server.
If I don’t have 100% control, I won’t take 100% responsibility for SEO results. No smart SEO will accept performance-based pricing without 100% control. And you won’t give it to me, unless you stared at the sun too long as a child.
Chasing one keyword or set of keywords is stupid
This only applies to the “I’ll pay you X when I rank Y” model. There are other performance-based models, which also suck. I’ll get to those in a minute.
Trying to rank for That One Keyword is fun. But it’s a terrible disservice to your company. If you pay attention to search visibility first and keywords second, you’ll get better results:
- You’ll move up in the rankings for lots of ‘long tail’ keywords – longer, less-competitive phrases.
- Those longer phrases, combined, will almost always generate more traffic, sooner, than any one super-competitive keyword.
- Those longer phrases will also bring you visitors with better, more focused intent. They’re later in the buying cycle, or more interested in additional information, or seriously weighing their voting options.
Also, rankings have become very difficult to measure. With the rise of personalized search results, two people can see two different rankings for the same term, on the same day, in the same city. Whose ranking is ‘right’?
Even worse, paying me for rankings just encourages me to be naughty. If you’re going to pay me $100,000 after 90 days in the #1 position for “bicycles”, then all I have to do is keep you there for 90 days. I can invest $40,000 in brokered links and some parasite hosting and use a few other dirty tricks. Bam. You get your rankings.
And bam, the moment your check clears, I take down the links and parasite pages to save some money. You plunge out of the rankings. Buh-bye.
Finally, paying for rankings encourages poor marketing practices. It tilts your entire strategy away from great storytelling and smart site development, towards “how many times can I repeat ‘diapers’ on this page?”.
So, what’s the right move? Not paying me for rankings.
Paying for traffic fails twice
Next option: You can pay me for every visit you get from organic search.
But it’s prone to the same short-term thinking that kills the pay-for-rankings model. If I do SEO now, you’ll see much of the boost months from now. That gives me very little incentive to work based on the traffic I generate, because even on a 12-month contract, I’ll basically get paid for 9 months’ work.
You can set up a contract that promises to pay me months after we stop working together. But after a few rough experiences there, I trust your company about as far as I can lob a 747.
The moment traffic really tops out, you’ll realize you’re paying me a ton of money. Your boss won’t care that it’s earning you 3x that much. All she’ll care about is that line item. I’ll become a gnawing annoyance in her belly, until she either rips up the contract or has me run over by a street sweeper.
Note I said your company. I trust you. But a year from now, when you leave, I’ll end up arguing with some shyster thinks his law degree from Billy Bob’s Online Law School gives him the right to tear the contract up, one niblet at a time.
But attribution is what really kills pay-for-traffic. You can pay me for every click over NN clicks from organic search results. That probably means you’ll end up paying me for a lot of clicks I didn’t generate, and missing a lot of clicks I did. It’s very difficult to precisely attribute clicks to specific efforts. Since the pay-for-traffic model depends on that precision, it’s an epic fail.
I do think this model has promise. Companies like Enquisite offering better and better measurement tools and pricing systems. At some point, if everyone decides to play nice, this system could be a fantastic way to price out contracts.
You’re not paying for a ranking, or traffic…
…You’re paying for a result, and a relationship. You’re for:
- Education: You’re going to learn a lot. Unless you clap your hands over your ears and yell “lalalalalala”. That won’t show up in traffic or rankings for a long time. But it’s still a huge benefit, and you’re gonna pay for it.
- Expertise: Guess what? I’ve spent 14 years learning SEO stuff. It’s really hard. And you can’t pay for that by tipping me 10% for a high ranking.
- Agility: One week rankings matter to you. The next week, some jerk posts a negative review of your company on a forum and it starts ranking #2 for your company name. D’oh. You need me to shift gears, fast, and help with the new challenge. Hate to tell you, but that kind of agility costs money – it requires a multi-talented team. If you want me to charge less, I can replace my team with guys like this.
- Safeguarding your brand: You want top rankings, but you also want to not be perceived as the biggest butthole on the internet. That, too, makes our work harder. But you can’t measure it.
- Results: When I say ‘results’ I mean an outcome that grows your business. Sometimes that’s straight-up sales or leads. More often it’s that, plus phone calls, plus other stuff that you can’t precisely measure. Nor will you ever be able to. Sorry. You won’t.
- Relationship: You’re paying for trust. #3 in this list only becomes a huge problem if you don’t trust me. If you don’t trust me, don’t work with me. Please. I’m begging you. Clients who don’t trust you are like psycho girlfriends/boyfriends: I shudder every time the phone rings.
What really works
The best models I’ve seen work one of three ways:
Retainer-based. Not always popular with clients, but provides the most flexibility to your agency. You pay a flat fee each month. They work their butts off and report back every week/month. Repeat. If they suck, you fire them later on.
Flat fee-based. Pay someone a flat fee for a super-deep one-time review and recommendations for your site. Then you either make the changes on your own or pay someone else to make those changes. This limits long term results a bit, since there’s no steady effort. But it also limits your risk. If you’re an in-house SEO, it’s a great option.
Percentage of revenue or pay per lead. This can work. It’s another performance approach but it’s tied directly to the company, not rankings. But I’ll only do it with 100% control of the site and marketing. Other internet marketing geeks may be more flexible. I’m not.
Compromise is best
I’ve also seen a mix of all three models, plus performance incentives, work best. For example: Pay a retainer, plus a bonus if X happens. Or pay a retainer plus a percentage of sales. Or something similar.
I do see systems like Enquisite’s, that provide a baseline for SEO pricing, as being essential in the next generation of SEO campaign pricing. No, you can’t use performance-based pricing as the long-term benchmark. But you can use it to figure out what your work should cost.
Whatever you do, build a model that’ll reward great performance while also giving your SEO pro an incentive to make an effort right now.
OK, you may leave your hate mail, comments about what a whiner I am, etc. below. Or blame Oilman for inspiring my post with his.
Ian Lurie is CEO and founder of Portent Inc. He's recorded training for Lynda.com, writes regularly for the Portent Blog and has been published on AllThingsD, Forbes.com and TechCrunch. Ian speaks at conferences around the world, including SearchLove, MozCon, SIC and ad:Tech. Follow him on Twitter at portentint.He also just published a book about strategy for services businesses: One Trick Ponies Get Shot, available on Kindle. Read More