• http://twitter.com/littlefishweb littlefishweb

    Nailed it! “SEO is not a one-and-done job. It is an on-going series of identifying problems, implementing corrective measures, examining real-world results, and refining those efforts by starting the loop all over again.” And you should win a Pulitzer.

  • http://twitter.com/jhuinink Jim Huinink

    I laughed, I cried, I had to tweet it. Thank you, Rick.

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

    RIck, this is exceptionally well-crafted. Kudos on capturing the heart of an issue that plagues all quality SEO shops. Indeed, paid search suffers some of the same challenges. We can identify all kinds of problems with a prospect’s account, but how much the prospect’s performance numbers will change from making those improvements is impossible to gauge precisely because even in ppc, user behavior, search volume, and competitor behavior are outside of our control.

  • http://www.facebook.com/mikezittel Michael Zittel

    Great to see other SEO’s have same dilemma. We do give “forecasts” as needed, but of course there are no guarantees. To create a legitimate forecast we base it on potential traffic, existing click, conversion data, etc. We also try to steer clients away from to much analytic data and keep them focused on ROI. It all depends on the client, their product, sales cycles, etc.

  • http://twitter.com/dccutler Dan Cutler

    Great post – I think that there’s definitely enough quality SEO data to produce fairly accurate projections and pretty reasonably accurate reports. Between all the kinds of data we have you can triangulate things pretty well so to speak.

    PPC is awesome in this respect. Managed well it produces instant credibility w/ clients and lends authority to the organic SEO part of your inbound strategy.

  • http://wordswordsseowords.com/ Christopher Skyi

    The bean counters want things on what’s called an “interval scale,” which is just a set of hard objective numbers. How many traffic accidents will occur in NYC today can be put on an interval scale and you can measure the variation over a year in terms of mean and standard deviation. If you have a great model of what causes traffic accidents, you can make a prediction that will fall on an interval scale.

    Most of the time, however, our understanding of the world is less precise, so all we can say is that if we improve traffic lights in some way, we’ll get “less” accidents, but we can’t say by how much. This is called an ordinal scale where you can rank something as more or less than something else, but you either can’t objective measure how much more (e.g., how much will your wife love you if you remember her birthday? She’ll love you more but you can’t measure exactly how much more — we don’t know or understand that scale) or you can’t predict how much more. In the case of SEO, we can measure how much more traffic objectively, but we can’t predict that.

    This gets tricky to explain, but a given SEO task will result in “more” traffic, a hard number, but you can’t know exactly how much more. Statistically you can, in principle, determine if it is really “more” because of the task and not just by chance, but that’s about it.

    So SEO isn’t uncertain. It is just that there is no “mapping” between a particular task and a specific increase in traffic (or if there is a mapping, we don’t understand enough about that aspect of the world to know what that mapping is). All we can say is “more” or “less,” “better” or “worse,” and maybe we can even say if it’s significantly better or worse or no difference given the right statistical test.

    User website satisfaction is clearly on a ordinal scale, e.g., if on a scale of 1 to 10 you rate a website 3 but I rate it 6, you can’t say I like it “twice” as much. All you can say is I like it significantly more than you, and the higher I rate it, the more certain you become that I like it more.

  • http://wordswordsseowords.com/ Christopher Skyi

    The bean counters want things on what’s called an “interval scale,” which is just a set of hard objective numbers. How many traffic accidents will occur in NYC today can be put on an interval scale and you can measure the variation over a year in terms of mean and standard deviation. If you have a great model of what causes traffic accidents, you can make a prediction that will fall on an interval scale.

    Most of the time, however, our understanding of the world is less precise, so all we can say is that if we improve traffic lights in some way, we’ll get “less” accidents, but we can’t say by how much. This is called an ordinal scale where you can rank something as more or less than something else, but you either can’t objective measure how much more (e.g., how much will your wife love you if you remember her birthday? She’ll love you more but you can’t measure exactly how much more — we don’t know or understand that scale) or you can’t predict how much more. In the case of SEO, we can measure how much more traffic objectively, but we can’t predict that.

    This gets tricky to explain, but a given SEO task will result in “more” traffic, a hard number, but you can’t know exactly how much more. Statistically you can, in principle, determine if it is really “more” because of the task and not just by chance, but that’s about it.

    So SEO isn’t uncertain. It is just that there is no “mapping” between a particular task and a specific increase in traffic (or if there is a mapping, we don’t understand enough about that aspect of the world to know what that mapping is). All we can say is “more” or “less,” “better” or “worse,” and maybe we can even say if it’s significantly better or worse or no difference given the right statistical test.

    User website satisfaction is clearly on a ordinal scale, e.g., if on a scale of 1 to 10 you rate a website 3 but I rate it 6, you can’t say I like it “twice” as much. All you can say is I like it significantly more than you, and the higher I rate it, the more certain you become that I like it more.

  • Alan

    Great post probably one of the best on searchengineland. The shame is that no SEO client will ever read this. If we could get the clients to read this kind of stuff it would make our jobs easier.

  • Alan

    Great post probably one of the best on searchengineland. The shame is that no SEO client will ever read this. If we could get the clients to read this kind of stuff it would make our jobs easier.

  • http://twitter.com/salyris Salyris Studios

    This is why you strategize, implement, measure, analyze the results and repeat!

  • DPM

    As someone who has been an SEO client, reading this article
    is exactly what concerns me about hiring an SEO firm. I completely understand
    the uncertainties that exist because you as a firm do not control Google and
    the other search engines. However, you are selling your service to “improve”
    your clients’ rankings, which is a very vague result in itself. Hard
    projections and estimates are clearly not possible as you describe, but without
    them how do you measure the success of an account? Simply by the fact that “it
    went up”? What if it only when up 2x when the real opportunity is 10x? I
    understand that you as a firm need caveats that you cannot predict how the
    search engines will react…but what if you fail to deliver? Or what happens if
    you under deliver? Most SEO firms are paid based on their time and not their
    results…and that to me is the problem with the SEO industry in general. SEO firms
    ultimately have no real accountability, and your article highlights the fact
    that you can’t fully be accountable because so much is out of your control…but
    you can take a shot and see what happens. That makes hiring an SEO firm very
    difficult for metrics and results driven organizations like the client you
    reference in the article…we can kind of count on you all to give some advice
    but ultimately we don’t know if it will work.

  • http://makethemclick.com.au/library Mark @ Make Them Click

    The answer to her question is actually quite easy if somebody is doing their job properly. That may be the marketing manager herself, the SEO company or somebody else entirely such as the business analyst.

    The answer of course lies in analytics. If a company has its analytics set up correctly it will know with a high degree of certainty how much business it gets from organic traffic.

    Running the data through some basic formulas in a spreadsheet will reveal how much extra revenue it will get from increasing its search engine performance.

    Using this data it can then set a budget for SEO and know what the likely return will be. It won’t be 100% but the beauty of analytics is that you can continually measure the performance and fine tune it as you go.

    It’s actually not that hard a question to deal with.

    Although as we all know many companies don’t have their analytics set up properly in the first place

  • Heather Baker

    Not sure I agree with the argument in this article. SEO, to me, is the process which gets your website ranked in the top pages of the search results.

    SEO, on it’s own, is not what guarantees the clicks and conversion. It is the brand message and the clients’ web content that achieves the clicks & conversions, which can only be achieved by teaming up with skilled brand managers, copywriter and maybe even a PR team.

    It’d be far more effective to throw this explanation at a client when asked about the expected results from an SEO campaign.

    This article isolates SEO from the rest of the essential marketing, communications and branding elements of a successful digital marketing campaign. If clients are lead towards taking this approach with SEO then they’re never going to get the results they desire.

    This, in my mind, is why it’s essential to hire an integrated marketing and communications agency who can develop a holistic strategy that achieves better results. It’s becoming more apparent that SEO, in it’s traditional form, is a dying art. It now needs to be integrated with PR, Social Media and even Video if clients’ want those extra clicks and conversions.

  • http://www.brickmarketing.com/ Nick Stamoulis

    “The hard but honest message to convey to clients is that there is
    uncertainty around the level of expected success that will be seen in an
    SEO campaign.”

    And I can understand why that would frustrate a
    client, especially one that has to report to the “bean counters.” It’s
    hard convincing someone to invest in something where you can’t give them
    hard and fast numbers upfront about what kind of impact their money is
    going to have. It’s the song and dance of the SEO company-client
    relationship.

  • http://twitter.com/Nick_Boyle Nick Boyle

    Rick, this is a great post – hats off for taking the time to put it together for us to read. Could not agree more with the ‘why there’s no certainty’ section. This post is just TOO good to not share – regardless of whether you agree with the points raised or not. It’s this sort of content that so many SEOs fail to create… they’re too busy churning out 15 low-quality pieces per week!

  • http://twitter.com/sharithurow sharithurow

    Amen! You are preaching to the choir!

    Search has always been a part of information architecture (Polar Bear book anyone?), and a subset of findability as a whole. I have never understood why user experience (UX) professionals seem to discount or diminish the value of search as part of the user experience. Shouldn’t the searcher experience be a positive one? Users are searchers.

    I have never guaranteed results, and it has cost me many leads and clients. But my methodology, when implemented properly, works like a charm. Most of my clients only implement a few recommendations and expect miracles.

    I think this article and topic was sorely needed. You have a permanent link from me, Rick. Nicely done!

  • http://www.facebook.com/nick.frazee Nick Frazee

    I normally do not comment but this is an award winning post in my book!! I had this conversation yesterday with an attorney client. Thank you for this post every new hire I have will have to read this before meeting with a client.

  • Derek Abbring

    This stems back to the saying I hear a lot “You never know what Google wants” If you hear this, run. Run far away. You should never hire a firm that doesn’t know if their services will work or what the results are that they can achieve.

    If they do not offer a guarantee of any kind, chances are they will only know enough about SEO to sell it, but then outsource the actual work load – which is why they can’t guarantee or predict results.

    This is where the issue comes, businesses looking to stay frugal then hire discount, unqualified SEO firms that outsource link builders & content writers.

    My firm guarantees our work, we predict results and we haven’t had any issues providing expected white hat results to date. Lots of hours on our end to hold up promises, but a lot of happy clients as an end result.

  • Derek Abbring

    This stems back to the saying I hear a lot “You never know what Google wants” If you hear this, run. Run far away. You should never hire a firm that doesn’t know if their services will work or what the results are that they can achieve.

    If they do not offer a guarantee of any kind, chances are they will only know enough about SEO to sell it, but then outsource the actual work load – which is why they can’t guarantee or predict results.

    This is where the issue comes, businesses looking to stay frugal then hire discount, unqualified SEO firms that outsource link builders & content writers.

    My firm guarantees our work, we predict results and we haven’t had any issues providing expected white hat results to date. Lots of hours on our end to hold up promises, but a lot of happy clients as an end result.

  • http://wordswordsseowords.com/ Christopher Skyi

    You want to see positive trends, i.e., more traffic over time and more keywords ranking. You can definitely and objectively measure progress using Google (or Bing) Webmaster tools and Google Analytics (or something similar). Google has gone out of it’s way to provide these progress measuring tools, and for free! If you want to go the extra mile, hire someone who also knows how to use Google Analytics or something similar to improve your website to keep more of the traffic your getting to stay on the site rather than just bounce away and to improve sales and leads for the current traffic your getting.

    Many small business owners got nailed by Panda and especially by Penguin and while they rightly blamed their SEO firm, really — they also had themselves to blame because they wanted those hard hard numbers, especially ranking numbers, e.g, “first page on Google Guaranteed!,” “Number one spot on page one Guaranteed!,” etc. when Google themselves have said that #1, or any specific position, cannot be guaranteed and that anyone promising that is selling SEO snake oil/black hat. But they fell for it anyway. And then paid a BIG price for those promised high rankings, if they even managed to obtain them in the first place.

    SEO not some kind of magic fairy dust sprinkled on the site or a bag of special “tricks” available to you but not your competitors, and it’s not something you can just hand over to a firm or consultant, at least in most cases. You’ll need to be involved, probably more then you think.

    The cold hard truth is that effective SEO is HARD work, it takes time and — it’s now expensive, but you definitely can measure progress. However, if you’ve only got $100-$200 per week or month to spend, or if you only want to invest for a few weeks or months, you’ll have to forget about increasing your traffic organically (unless you’re willing to do all the hard work yourself) and instead spend the money on adwords. Spending a little money on adwords will give you far better value then spending that money on half-hearted, or worse, bad or black hat SEO. In short, SEO is a lot like life — success takes hard work, time, smarts, and, like life, there are no guarantees, but if you’re really willing to go for it anyway, really commit and persist, you will find success.

  • http://wordswordsseowords.com/ Christopher Skyi

    I agree. This is definitely the future. The better integrated SEO is into the rest of a company’s marketing efforts, the more effective it will be. At minimum companies need to be doing SEO along with Google Analytics (or something similar) for feedback and for website improvement.

  • http://wordswordsseowords.com/ Christopher Skyi

    I’m not sure what sort of traffic/ranking guarantee your company offers, but all I can do is point you and others to Google.com Inc. has to say about that: http://www.youtube.com/watch?v=a6zGjcf6Snw

  • http://www.technostall.com/ Chankey Pathak

    An awesomely written article. Loved it, especially this paragraph:

    “If the world were as simple as “do X to get Y,” all of us would be rich, all products would be great, all marketing efforts would be perfectly successful, and, of course, this column would win the Pulitzer Prize.”

  • DPM

    100% agree that the progress, impacts, and results are all measurable, especially with the tools you mention. Unfortunately not all SEO firms are good at the data analysis portion either but that’s a different topic. Regardless, we have indeed committed our own internal resources to doing the “hard work” that’s needed and we’re bringing in SEO advisers/consultants when we feel like we need some advice. This formula has worked better for our company. Content generation is the easy part for our site…the biggest challenge now is around link building…and given the uncertainty around links, Penguin, etc, we don’t trust anyone outside of our organization to touch this at the moment.

  • http://wordswordsseowords.com/ Christopher Skyi

    ” we don’t trust anyone outside of our organization to touch this at the moment.” I think that’s very smart. I’m almost paranoid about everything I do. No way would I trust anyone else. Bringing Google Analytics into the mix is now a must for feedback about what you’re doing — ranking and traffic are the end result of doing some very positive things, things you would do, ideally, even if you don’t know about rankings, links, ect. In fact, no SEO is far better than too much or bad SEO. Right now everyone out there who is working on their site and knows nothing about SEO? They’re the chosen people! :)