Our bad experience with a vendor known as Yoast, in a perverse case of synchronicity, illustrates our theme of the crapification of everything. Yoast’s supposedly customized report to us is one of the shoddiest, error-filled documents I have ever encountered in my thirty-plus years as a professional. This Yoast SEO Review also provides a window into the power that Google has over Internet-based businesses. Plus we think you’ll find this to be an entertaining shaggy dog story, even though it was not at all fun for Team NC. (Busy techies who just want to see how wretchedly bad the Yoast SEO Review was, follow this link to see our discussion of the pervasive errors in their report).
By way of background, SEO optimization is “search engine optimization.” Since Google is the big kahuna in this space, SEO optimization is entirely about pleasing Google, so that your site come up well in its searches. I haven’t been eager to pursue this idea (after all, this site has done well by being feral) but after our redesign, I thought it might we worth considering.
Off I went to Yoast, an SEO consultancy which is also known for its Yoast WordPress SEO plugin. Yoast offers several level of review, but they all promise to be specific to your site, and they profess to do more than just look at SEO issues, even though on the intake form, we made clear that SEO was our interest. Even before we signed up, it was clear that Yoast makes its real money doing what amounts to consulting work for large sites (and of course, the Yoast WordPress SEO plugin). Thus its reviews for smaller sites are essentially repurposing the expertise they gained on big clients for smaller ones. In theory, this could be a good by-product business.
In our practical experience, not so much. The “review” Yoast gave our site was full of cut and paste boilerplate, and worse: chock full of flat out incorrect statements about our site and our business. Many of the recommendations weren’t merely useless; some would have damaged the site’s functioning or our business had we been foolish enough to implement them.
And even worse for prospective clients: Yoast has dropped its work standards even as it is massively upping its prices. We got in under the wire and paid their old price of €849, which was roughly $1100. They have since more than doubled their price for this type of report to $2500. Even Kristin, our designer and code jockey, who recommend Yoast told us that’s too much for the better quality report another client of hers got over a year ago (as many of their recommendations were not well spelled out, even after getting their answers to questions. They found they had to do far to much in the way of guesswork and experimentation to implement the Yoast advice).
But aside from the utterly embarrassing errors, which we detail below, Yoast didn’t bother to understand our business at all. As Richard Smith observed, “At one point he did actually notice that NC was a blog, but he didn’t really seem to make full use of that observation.” And there may be a reason for that. Our experience suggests that SEO optimization is not a great fit for financial and potentially also for political sites. The result is like trying to grow blocky, thick-skinned tomatoes because they’ll ship better, which has in fact become the industry norm, to the ruination of the taste of tomatoes. By contrast, NC’s discerning readership cares about matters of taste, and so conforming to the mono-culturist demands of Yoast would undermine many elements our readers find attractive.
Indeed, the idea that their could be such a thing as tradeoffs between SEO and business needs seemed to be an idea that Yoast was simply unable to grasp. As we detail, proof of our thesis comes via Business Insider, the most wildly successful site in our space. It started up in March 2008 and now gets more pageviews than the Wall Street Journal. Business Insider clearly, and deliberately, flouts some of the supposedly ironcast rules that Yoast insisted we follow. Yoast seemed unwilling to consider, for instance, that making people aware of your site and your posts isn’t sufficient; you also need to get them to click through and then stay. We happen to be particularly good at that, yet Yoast treated us as know-nothings who should swallow his advice without question.
Now to the slugfest.
We sent Yoast a letter and a spreadsheet detailing the errors (since they said they want feedback) and asked for a redo or a refund. In the interest of full disclosure, we are embedding that letter and the review itself at the end of the post. But our exchanges below do a good job of covering the ground of our missive, and will be easier for most readers to follow.
Since we had some back and forth, we’ll put Yoast’s responses, which came from Joost, the head of this small company, in italics.
After we sent our critique (and odds are high that NC team collectively spent more time on the letter and spreadsheet than Yoast spent on its report), we got a haughty reply that basically told us that since Yoast got hired by big clients, we were idiots and could take a hike. But that’s clearly a spurious argument. If a heart surgeon leaves a scalpel in a patient’s chest, how many successful operation he did in the past is irrelevant. He clearly made a hash of this one. In other words, this was a flat-out refusal to take responsibility for their deficient work product.
March 10, 4:44 PM EDT:
I felt obliged, seeing such an intensely negative response, to dive in and see what was up. First let me say that I appreciate that you took the time to write your long response. Regardless of whether we agree on a lot of topics, I think we don’t, I appreciate that you were so angered by the whole thing that you took the time to write this piece.
When we say we’re blunt we don’t mean to offend you, I must say that I got the feeling you were out to offend us, and with that I take issue. Now had you found a lot of things I could agree with, I would have understood your enormous anger, but I honestly don’t.
You make enormous assumptions about what we saw and didn’t see. You also seem to think that we don’t know the first thing about publishing on the web. You begin with all sorts of claims about McKinsey etc. Let me tell you my background and why I think you’re absolutely wrong in many, many areas. I work as an SEO and growth consultants to several big companies you’ll have heard of. To name a few we currently work for:
– the Guardian, for whom I recently completed their migration from .co.uk to .com, as evidenced by this blog post;
– Disney, where I recently had the pleasure of speaking to their entire worldwide SEO team;
– RTL in the Netherlands, one of Europe’s biggest TV companies;
– etc. etc.
And these are only the publishers, I’m leaving out current clients such as eBay and Facebook as well as past clients such as The Next Web and other huge online portals. Now, why do I tell you all this? Because I want you to believe that I both know what I’m doing and I’ve had quite good results with many companies leading me to where I am and how I’ve trained my team. I honestly think that you’re making some huge mistakes in how you’re looking at our report. While there probably are mistakes in there, there always are, almost NONE of your remarks ring true to me.
To me, the two most important chapters would be:
-Ranking on keywords.
While this was a short chapter, it’s paramount to your future success on the web that you understand the thinking behind this. But let me combine it with two others you had objections to:
– Page Titles
– Activate WordPress SEO/Check WordPress SEO
You see, if you’re not willing to optimize your content, you might as well not have ordered our review. You claim to know journalism, well let me tell you this: every major newspaper I’ve worked with over the last 10 years of doing this, and I’ve worked with quite a few, has dedicated staff for content optimization. The Guardian has 24 hour SEO coverage, nothing important gets published without an SEO having looked at it. If you’re time constrained, writing LESS and optimizing it MORE is the solution, not the other way round. You called us insane, to be honest, if that truly is what you think, please go back to publishing on paper. You might survive on the web because your content is good (I’m not contesting that)
Now let me go over some other things:
You’re unwilling to use headings, even though these could be made to look like any of the other ones, you’re not agreeing on Jetpack etc. It sounds to me as though your developer pushed back instead of reading and listening.
Combine and minify CSS and JS files.
“We have only one CSS file.” No. You don’t. You’ve got 5. Open your website, hit view source, search for .css and you’ll see I’m right. You make it very hard for me to take your critique seriously if several things you state are factually incorrect.
In all, I don’t like being this aggressive, so let’s see how we can both change our approach to this.
I’ll be honest: I’m not willing to give you a refund, nor am I willing to re-do the report. What we wrote down was the right thing to say. I am willing to shed more light on why I’m so passionate about this and why I think you’re so wrong. I’ve helped many many news portals gain many unique visitors a day and I want our customers to achieve greatness.
“I’m highly confident that I have been in the professional services business longer and have dealt with far more professional services companies (other management consultants, top and mid tier law firms, accounting firms, IT consultants) than you have”
while it may be true because you’re older, is actually petty. Especially if you knew who we worked for and what we do. I urge you to re-read the report with an open mind (although I accept that after this exchange of emails that might be hard), and consider that we know very well what we do.
Joost de Valk
CEO at Yoast
This was the last sort of response I expected, and I dashed off a reply at 8:21 PM the same day:
You team did grossly deficient work and I see you are refusing to take responsibility for it. You made specific promises on your website and did not deliver on them.
The work you did for others is irrelevant. I see you have made NO specific responses to the numerous deficiencies I pointed out. All you sent is “I do work for big clients, ergo my work is OK.” Sorry, that cuts no ice. You apparently couldn’t even be bothered to check my complaints in any detail to see what a shoddy job was done.
I’m publishing my letter and letting the world see how you engage in bait and switch with small clients.
That led to politer in form but every bit as remiss in substance reply from Joost on March 11 at 2:58 PM:
I mentioned our clients in response to your very specific accusations. Now you see the work we did for others is irrelevant, even when it was you who brought up YOUR work for others. You’re unhappy with what you got, I’m sorry for that. But I in fact took (and still take) full responsibility for what we did, because it was good work. I have checked your complaints in details, I’ve even shown you that some of them are just entirely untrue. These aren’t opinions, I can count how many stylesheets you have for instance, and you apparently can’t. Further arguing doesn’t seem to make sense, so I won’t.
We’ll be here if you have trouble implementing some of our advice, or have trouble understanding specific bits. Until then, I wish you all the best, both in life and with your site.
Joost de Valk
CEO at Yoast
We sent this message March 12 at 1:02 AM:
Thank your for your speedy reply.
However, with all due respect, your message to me of March 10 and your message below are not consistent with taking responsibility for your work. Merely asserting that you have done so does not make it true.
In my letter, I identified numerous, indeed, pervasive errors and tacit assumptions about our site, www.nakedcapitalism.com. That is conclusive evidence that you failed to do the review of our site that was promised. You have failed to denied any of the issues we have raised about the pervasive deficiencies in your work.
Moreover, punting on the necessary effort to do an adequate review led to either irrelevant or, even worse, potentially damaging recommendations in your deliverable.
As we will demonstrate, the two issues you attempted to address merely confirm that your firm did even begin to do the investigation necessary to provide the bespoke review your promised on your website. Hence, you have not delivered the service we paid for.
From the very start of your report, it is blindingly clear that:
– You could not be bothered to understand Naked Capitalism’s business at all, even though your promised to do that on the intake form on your website
– Substantial portions of the report are canned with at best token customization
You thus owe us a full refund or must provide work that was stipulated but has yet to be delivered.
I have already identified the following inaccurate statements and tacit assumptions about our site. These errors and misguided assumptions in turn produced irrelevant and, even worse, counterproductive recommendations:
1. Assuming we have a high bounce rate when we don’t;
2. Assuming we are selling a product (this false assumption is pervasive and repeated in various ways throughout the report, with cut-and-paste text that was not adequately edited, such as the clearly canned discussion under Testimonials: ” “bragging about you and your product…asking some customers…might convince him to make a commitment by subscribing to your newsletter or buying your product.” We do not sell products. We do not have customers, we have readers. We do not sell a newsletter;
3. Assuming we have a conversion problem (contradicted by our high # of page views per visit and high time spent on site per visit). Did you miss that addressing this non-problem was among your highest priority recommendations?
4. Not recognizing the site is a financial journalistic/analytical site, which is reflected in various ways, such as #2 above, and in your recommendation that we have testimonials and have introductory content (two separate poor ideas resulting from failure to examine our site enough to understand what business we are in, a remarkable oversight, particularly since I said in your intake form that we are a top financial and economics blog);
5. Making aesthetic calls that are debatable at best (and with which our readership, which includes numerous web designers, disagrees) as if they were gospel (for example, the claim we needed more whitespace for headers to be identifiable);
6. Assuming that we sell our own advertising and serve our own ads. This is a major miss and as with all your other errors, was the result of a failure to investigate (it was easy to ascertain by multiple means, by seeing how the site loaded and by looking in our backstage, to which you had access, at your request, as an administrator). The resulting off-base assumption produced numerous other errors as we discussed in the letter, such as number 7:
7. Assuming we did not have expert advice on our ad sizes and locations. We’ve had extensive input from our ad service, and its professionals have considerable expertise by virtue of having sold ads to large agencies for 15 years (as well as a clear monetary incentive to get this right);
8 Assuming our remnant ads were not well optimized (we get a very good CPM on them) and that we could sell ads directly (a remarkably uninformed suggestion; sites in our space do much better in terms of CPM by being bundled with sites with similar demographics and sold as a bigger ad buy to much larger advertisers);
9. Incorrectly assuming simple keyword searches are germane to our site. In addition, as we elaborated on our spreadsheet which we previously supplied to you as an attachment, implementing your strategy would be extremely costly in terms of ongoing maintenance (the extremely large range of topics we would need to cover, as opposed to “Florida housing” and “Florida real estate” and for a real estate firm in Florida) with vastly lower payoff than for a product firm. Again, your remarks here looks like cut-and-paste advice intended for product-oriented sites;
10. Assuming readers prefer excerpts to the ability to scan more posts faster (this based on specific feedback from our readers);
11. Failing to recognize that the two main authors on the site use pseudonyms, as do many of our other authors, and that this practice is prevalent in the financial blogosphere. We thus cannot disclose more about most of our writers. That makes your author page advice moot;
12. Assuming that we have not experimented with both Google search and WP search for internal searches. We’ve already picked the one we find works best for us and our readers;
13. Making incorrect assumptions about the number and preferences of our email subscribers;
14. Presuming that our site has child-of-child pages (for instance, Shoes -> Women’s -> Sandals -> [price]) when it was trivial to ascertain that we are not a product site, and readers are only at most one level below the site landing page. That led to an irrelevant, cut-and-paste breadcrumbs recommendation;
15. Missing that we serve our mobile version through the third-party service Onswipe;
16. Missing that we allow readers to opt out of Onswipe, and that the overwhelming majority do;
17. Failing to investigate how the desktop version of our site renders on smaller devices;
18. Failure to consider that short titles are counterproductive for many of our most important audiences, such as journalists and busy investors, who will not read an article with a vague general headline (unless you are Nobel Prize winner Paul Krugman). I just looked at the wildly successful financial blog, Business Insider, which having started less than six years ago, now gets more pageviews than the Wall Street Journal. Of the 20 most recent posts in its RSS reader, 18 have headlines longer than 60 characters. Moreover, some chose to add detail in ways that specifically push them over your supposed ironclad rule: “The DOW FALLS: Here’s What You Need to Know (DJIA, SPY, SPX, QQQ).” We are also close to the publishers of another high traffic site, Alternet, and they experiment obsessively with headlines. They too use longer headlines due to the resulting much higher click-through rate. In other words, your cookie-cutter advice fails to consider: getting someone to see a headline is insufficient, you need to induce them to click on it; and that our audience consists of particularly voracious news consumers who need to be told why an article is different from the 5-10 others they’ve stumbled across on the same topic to motivate them to read ours.
19. Describing our modest blogroll as enormous;
20. Treating our site speed issue as if it was native, as opposed due to ads coming from a third-party server;
21. Failing to test site speed on the desktop version of our site on small devices;
22. Failing to recognize that we have already combined and minified, and taking that recommendation further would break the site;
23. Cutting and pasting your Jetpack discussion from another review, and failing to edit out the blatantly incorrect “since you don’t have comments on most of your site”;
24. Related to 23, incorrectly stating we have hovercards enabled and suggesting we turn them off. Had any actual review of our site been done, you would have seen that hovercards were not enabled at the plugin level, in Jetpack;
25. Assuming all the stat packages are for our use, as opposed to dictated by our advertisers (we have only one for our purposes);
26. Incorrectly assuming that because we have had a lot of authors over time, that we have a complex managerial problem. Had you bothered looking in our backstage, you would have seen that on any one day, only one person is running the site, and the exceptions are rare days when two people split the load;.
27. Saying we don’t have a sitemap when we do.
We also pointed out that your report was unclear or unimplementable given our business on the following issues:
1. The use of schema
2. What you meant by “pages” in Meta Description
Your response of March 10 simply provides more evidence that the investigation of our site, and your reading of our letter, was superficial at best. Of 30 issues above, you addressed all of two, and in a way that frankly only dug your hole deeper.
You reply provides more proof that you did not read what we wrote with any degree of attention and thus failed to provide anything approaching a bona fide response.
Headers. On the matter of headers, you attributed my comments about headers as “pushback from my designer”. That confirms you did not read my letter carefully. First, we conceded the <h1> header was a glitch, something that was never intended to be part of the template (one of the h1 headers should be rendering as an image, as we explained, and not as a header). So pray tell, how was that “pushback”?
Second, we also raised the issue that your concern about duplicate <h1> headers could have been dated, since Google Webmaster Tools removed that check in an update of two or so months ago, suggesting they have likely devalued this as something they care about in site rankings/authority. You failed to respond to this query.
The third reading comprehension failure was attributing the objections raised to my designer, and not bothering to understand them. The objections were clearly presented as mine and those of my fellow authors. We are not about to use h2 or h3 headers within the text of a post, which is under the control of authors and editors. We have a following among policy-makers and need to use the visual vocabulary that orthodox for that space and that is not large, magazine-style headers.This is based on the fact that the heaviest heading weight we can use for a policy audience is a simple bolded text of the same size as the main text. At most, that could be a h4, making it of limited SEO value. We use h2 and h3 for more prominent template elements. The visual differentiation of larger size text is important there given all the ads.
Your brush off here shows you didn’t bother to read or understand our objections.
CSS. On the matter of CSS, your efforts at a “gotcha” backfired. I don’t see how anyone can take anyone seriously who looks at “View Source” and fails to comprehend the information it presents.
“We” have only one CSS file. “We” means the the code we design and control. We do have only one CSS file in our template. There are CSS files from plugins, Google fonts, etc. that we have tried to combine using plugins like WP Minify and CSS Combine and found the site to break. We cannot manually combine these files, as you know. Our only recourse is to leave as-is and to periodically check new combining plugins to see if they work for us.
You can see that we indeed have only one CSS file in our template from “View Source.” But you either failed to do more than a cursory look or chose to misrepresent what you saw.
And your “gotcha” also appears to be a device to distract attention from a far more serious bit of bad advice you gave, that of taking minification and combining further. As we said, and you have not disputed, had we implemented your advice, it would break the site.
You raised a final objection to our point about the failure to qualify customers. As for your “if you don’t have a lot of resources to do SEO on an ongoing basis, you should never have hired us,” that is YOUR failure, not ours. In my letter, I stressed that it is incumbent upon service providers to qualify their customers to see if they can do a decent job. For instance, if someone goes to a trainer and says, “I want to lose 100 lbs. in 2 months,” an honest professional does not take their money. He says he can’t do what they want and indeed no one could do what they want. If he still wants their business, he needs to talk to them to see if he can deliver results that might fit their limitations and needs.
I told you on your itty bitty “don’t tell us anything” intake form that our site was severely resource constrained. At that point, you had taken our payment but had not started the report. The onus was on you to ask what that meant if you didn’t understand what “severely resource constrained” meant. Saying we should have known whether you could serve us well, given your refusal to do any sort of intake assessment whatsoever, is tantamount to asking us to do self diagnosis. We are in no position to do that.
Conclusion. Based on your responses thus far, it is unlikely to be productive to continue this conversation. Your actions in the last two messages are inconsistent with your statements. You offered to “be here” in case we had questions. However, your continuing unwillingness to do the basic investigation required to understand our site, as reflected in your cursory, dismissive, straw man answers to our letter, confirm that you refuse to do the work you promised you’d execute for us.
(Techies, click here to return to the beginning.)
And for your further delectation, here is our letter and the Yoast report. Don’t waste your money on them.
UPDATE: Lambert here: We received a final email communication from Yoast. Since it adds nothing new, I won’t bore you with it. A reader summarizes:
He just says no point in discussing further he won’t refund and he did a good job and he knows your business better than you do.
He says we are technical idiots and he would love to explain things to us so we understand.
I’m paraphrasing. A little.
It’s like Team Yoast is bucking for coverage in an HBR case study entitled “Exactly How Not To Handle an Irate Customer.”
Yoast relented and provided a full refund, so his report to us has been removed.