Dear Team Yoast,
I received your review of my website, Naked Capitalism on
March 4. One of the things you mentioned in it ...
solicit input from them. For instance, we went through
five drafts of our redesign and got extensive reader
feedback on ea...
refund, since I have little confidence that a new report
would be much of an improvement. Again, you did not give
me what ...
Embarrassing Mistakes in the Report
Your cookie-cutter approach in Section 2 is clearly not
designed for journalistic/anal...
a day managing relationships with them and readers. This
is another concept that appears to be alien to how you
think abou...
404 pages. Not adequately investigated. Relevant only
if readers get a lot of 404 errors. You appear not to
have used Webm...
scan quickly and not have to go through multiple pages
(which is what results from including extracts).
Author pages. Lack...
Schema.org. This is one of the few potential high impact
recommendations, and you punted, giving us two canned
paragraphs ...
functionality of being able to check for duplicate
headers, suggesting that Google no longer cares
about this. If our assu...
schedule (which you could have seen had you
investigated) I have time only to auto-tweet headlines.
Again, a more, not les...
done work for finance, economics, or political blogs.
Thenumber of sites in my blogroll is low compared to
norms for finan...
Combine and minify CSS and JS files. Failure to
investigate. As we’ve mentioned repeatedly, we run ads
that come from a th...
WordPress stats are necessary in addition to that for
our own use.
Edit Flow. You didn’t bother understanding how we run
t...
recommendations 1-25, we cannot implement most of them
and the ones we can are ones you deemed comparatively
unimportant. ...
of 14

Naked capitalismyoastresponse

Published on: Mar 3, 2016
Source: www.slideshare.net


Transcripts - Naked capitalismyoastresponse

  • 1. Dear Team Yoast, I received your review of my website, Naked Capitalism on March 4. One of the things you mentioned in it was that you were being blunt. Let me respond in kind. Your review was not what you promised on your website. You repeatedly indicated that I would receivea review of “your site,” meaning that it would be based on a specific and detailed examination. Your site also stressed your experience in performing SEO reviews, (600 reviews of customers similarly situated to me, plus more in-depth work which much larger concerns). Therefore what I was sold was a review that would be based on understanding my site, performed by experienced, knowledgeable personnel. Instead, what I got was an embarrassment that looks to have been produced by a new hire mechanically following a checklist. It contains numerous statements about my site that are flatly inaccurate. The level of errors reveals a failure to make an adequate investigationof my site, a failure to understand my revenue model and audience and even worse, demonstrates a lack of experience with sites that are journalistic in nature. Your report follows a “check the box” format, and wastes considerable space on misguided and irrelevant recommendations that I did not ask for. Only a comparatively small portion of this report addresses what I was interested in, SEO. Most of those cannot be implemented due to constraints we identified on your intake form. The other bizarre unstated assumption underlying your report is that we don’t understand our readers. We have an extremely large, engaged, and highly vocal readership. They tell us what they want and don’t want, in our comments section and via e-mail, even about trivial annoyances. We also go to considerable lengths to
  • 2. solicit input from them. For instance, we went through five drafts of our redesign and got extensive reader feedback on each one before going live. And they pay us back, literally. A significant portion of our site’s revenues comes via reader donations, a fact that you appear to have missed completely in your analysis (as in keeping our current loyal readers happy is of paramount importance). The result, almost without exception, is that the recommendations based on these erroneous assumptions are misguided, unimplementable, and sometimes counterproductive. The few useful suggestion you have are buried among ones that are filler. Even worse, for these very few, useful, potentially high impact ones, you don’t provide remotely enough guidance for us to be able to implement them. So pray tell, what were we paying for? I base this view on the fact that I have been in the management consulting business since 1983, starting at McKinsey, and a considerable amount of my work has involved young technology-related ventures. I have seen a large amount of professional services analyses provided to me by clients and contacts. Your report is one of the worst I have ever encountered. And this is not just my opinion. I have shared it so far with 4 individuals, each of them IT professionals, each with over 25 years of experience in settings that would have them seeing third-party professional work, and all gave very bad grades to this report. I am thus highly confident that if I were to publish it on my blog, with this letter, that my and their criticisms of it would be seen as accurate by third parties. I will detail the considerable shortcomings with your report in due course. However, I expect one of two remedies: either a full refund or a completely new version of this report, based on a proper examination of my site, focused solely on SEO issues. I prefer a
  • 3. refund, since I have little confidence that a new report would be much of an improvement. Again, you did not give me what you promised and what I paid for. Sales Bait and Switch; Discouraging Clients from Providing Information Your intake form actively discourages clients from providing much information. Your e-mail with the link had the header “Two Intake Questions for our Review of Naked Capitalism.” The form stated: “Are there any specific things we need to know about your site? (Please keep it short, we'll read your site too.)” It had a remarkably diminutive box for putting in the answer, not even capable of displaying a reasonable-length sentence, making it virtually impossible to say anything meaningful. Thus your intake process actively discourages clients from providing you with information, and you are explicit about not wanting to hear much from the client. You instead strongly signaled that you wanted to do your own investigation, and made the unusually intrusive request of asking for an admin login, which I supplied. That approach would have fine if you were willing to do your own work or contact clients during the review with questions but neither happened in my case. It was only when I saw the intake form that I had any clue you intended to cover topics you had not mentioned as part of the site review and that I was not interested in (usability, conversion, site speed). (I signed up when significant numbers of your pages were down, including ones related to your site reviews, and your records will show I used the intake form for your consulting projects to ask if you had stopped doing site reviews). You market yourselves as an SEO site, I had gotten a referral from someone who had gotten an SEO review (by Joost, not a new hire) and was happy with the result. I never thought when I committed to the review that I would be paying for irrelevant advice. I checked the SEO box.
  • 4. Embarrassing Mistakes in the Report Your cookie-cutter approach in Section 2 is clearly not designed for journalistic/analytical sites like mine. You start with a boilerplate discussion of bounce rates, and presume that is a problem without having asked for or looked at statistics on what my bounce rate actually is. My bounce rate is 42%, which I understand considered to be good, particularly since my site’s name is Naked Capitalism. I will always have a higher bounce rate than other sites in my space because some people who visit the site will be looking for porn. And you’ve got lots of canned lingo that you couldn’t be bothered to edit to hide that fact. For instance: “Your product or service can be the greatest thing on earth.” I’m not selling products or services, I’m providing information. Conversions. You bizarrely assume we have a conversion problem in the absence of evidence. You make suggestions like a call to action for "conversions" but what a "conversion" means to us is simply that visitors click on links to see posts and comment on them, and we have a ton of links in their faces to click on already. We have a much higher than average number of page views per visit (3 pages, even with our bounce rate) and a very good average time per visit (6 minutes, which is almost unheard of) so your assumption that we have a conversion problem is counterfactual. We do extremely well when we get visitors. What we need is more of them. Testimonials. Testimonials are simply not done in the information/journalistic space. Please go look at any major newspaper or independent blog or any of the other tops sites in any of the spaces in which I compete (economics, finance, politics). Testimonials come via being quoted on and linked to at other sites. I work in a very active ecosystem (all the top bloggers know each other personally, and are known by top financial media reporters and columnists), and easily spend 2-3 hours
  • 5. a day managing relationships with them and readers. This is another concept that appears to be alien to how you think about sites. Whitespace. Boilerplate and incorrect. Your claim is readers can’t tell when the article begin and end. The change in type size and color with new article titles is a clear and obvious marker. Banners. You are out of your depth on advertising. You did not bother to find out, which you could have by either asking me, watching the site loading carefully or by looking in our backstage, that all of our ads are served by a third party, Investing Channel. You are probably seeing more remnant ads than most readers by virtue of being located in the Netherlands (many of our ad campaigns have geographic limits, which can be as specific as New York City area). Investing Channel is run by professionals who have been selling Internet ads since 1999 and place ads with top ad agencies. They provided extensive advice in our recent site redesign on the placement and design of advertising based on what is optimal in terms of what advertisers want, and when appropriate, reader response. They are also QUITE aware of the tradeoff between too many ads and loss of pageviews (and the risk is not as you claim bounces but adblockers). *** I am going to start being more terse here and will skip over some items in the interest of keeping this missive to letter rather than book length. However, just because we are silent on anything that appears subsequently in our response does not mean that we do not have objections to it or that your suggestions are based on inadequate investigation. ***
  • 6. 404 pages. Not adequately investigated. Relevant only if readers get a lot of 404 errors. You appear not to have used Webmaster Tools and instead gave us a canned recommendation. It is certainly a good idea to have an informative 404 page, but you should have looked to see how material it was for us to give us a proper sense of priorities. We have no broken internal links and virtually all of our 404 are served to Google bots or junk sites. This is a “nice to do when you can get around to it” rather than critical. Ranking on keywords. Irrelevant, reflects lack of understanding of news/analytical sites. We don’t care about keywords. We don’t sell a product. We would not expect to rank or be found using the word “capitalism” or any of the other major topics we write about. They are enormous topics and much higher traffic general news sites would clearly outrank us. We would expect to rank ONLY on more targeted searches. Moreover, to write a keyword page, we’d need to write that (and 100 other pages) and keep updating it whenever thinking or news on that changed. It’s double work and as indicated, this is just not happening, any extra work would result in a reduction of the number of posts we can publish, which will very quickly kill our business. Introductory content. Inappropriate. People come to our site for news and they want to get it quickly. They do not any more visual clutter (and we already have a lot thanks to the ads) to distract them from the meat, which is our news-driven content. Excerpts on Archive pages. We have extensive feedback from readers on this and you are incorrect. The only people who use Archives pages are regular readers, and they have asked for a “see a month on a page version, headers only” and provided models from other sites they liked. They want something they can
  • 7. scan quickly and not have to go through multiple pages (which is what results from including extracts). Author pages. Lack of basic research into our site. Our two lead writers and some of our other authors are pseudonymous and will not provide “proper biographies”. I have gotten death threats and thus am not willing to provide much detail. Moreover, we are not willing to expose ourselves more to NSA and private surveillance by providing any more information than we already do to Google via a Google Profile. And none of our major competitors use avatars, so we are not at a disadvantage by not using them. Taxonomies. False assumption. We have not been not adding more categories, as you incorrectly indicate; we’ve been at this level for three years. The only two we’ve added are “surveillance state” as a result of the Snowden revelations and “payment systems, and both are legitimately new topics for us. And I don’t see any value in a higher level categorization. It’s useless to readers and there is no way I see to create one that is MECE (mutually exclusive and collectively exhaustive). Internal Search. False assumption. We used Google search for years. For our purposes, WP search is better if still flawed. And if forced to have one or the other, we and readers WANT results by date, not relevance. Email. Factually incorrect and false assumptions. We do offer email subscriptions. More important, we have over 7000 subscribers, making your Mailchip recommendation a poor choice. You further assume that readers might want something other than daily. They don’t. No one wants a day old or week old newspaper. We have never once had a reader ask for anything other than a daily email. Aside from the fact that our content is news-driven, a second reason people want e-mails daily is our extremely popular Antidote du Jour.
  • 8. Schema.org. This is one of the few potential high impact recommendations, and you punted, giving us two canned paragraphs that amount to handwaving. Schema.org is notoriously user-unfriendly even to experienced tech professionals. This is where we should have gotten the benefit of your expertise, in how to implement schema, but you failed to provide it. Breadcrumbs. Inappropriate, and reflects failure to investigate the site. This is boilerplate that should have been deleted. Our site has only two levels: the landing page and individual post or header bar pages, that are only one level below the home page. There’s no point in breadcrumbs for a site with this flat a structure. Responsive website. Failure to investigate. Your reviewer missed that our mobile versions are served by Onswipe, a third-party which we do not control. We have asked Onswipe to make the choice of “visit desktop site” (upper left) persist for 30 days, and the overwhelming majority of our readers are doing that (Onswipe provides marginally higher revenues from the readers who do tolerate it, but we may ditch it anyhow). You evidently did not look at how the desktop version renders on the iPad, and our iPad users (including one of our lead writers, who even posts from his iPad) like how it renders. Heading Structure. Failure to make inquiries and possibly dated advice. We are not about to use headers withinposts. We do serious policy work, as we did tell you in our intake form. Law firm, Congressional reports, inspector general reports, think tank documents use only simple bolded text to indicate sections within document. We need to emulate those forms to signal that we understand the visual vocabulary of serious policy discourse. As for the “there should only be one <h1> per page,” while it’s not hard to clear up (our logo showing as a <h1> is some sort of anomaly), Google’s Webmaster Tools in the last two months or so eliminate the
  • 9. functionality of being able to check for duplicate headers, suggesting that Google no longer cares about this. If our assumption is correct, your remark is dated and suggests we are not getting current advice, which would be a major failure given how often Google changes its search priorities. Page Titles. This recommendation is insane. Go look at any news site, or the most successful financial blogs, our direct competitors. They almost never have less than 60 character headlines. People are not going to waste their time clicking on an article unless they have an idea why they should do so. Moreover, because we are in finance, signaling what a post is about can inherently take a lot of characters because the subject matter is complex. For instance, there is a new scandal about a company called Ocwen where the New York State banking regulator Benjamin Lawsky is after them for how many mortgage servicing rights they’ve bought recently, which has led them to do a bad job at servicing. “Mortgage servicing rights” + Ocwen + “New York’s banking regulator Lawsky” have to be in the headline, and we’re already up to almost 60 characters before saying what the issue is. Most of our posts are like that, in that they cover material where you need to use specific terminology in the headline. A second reason this recommendation is insane is one of our most important audiences is journalists, who serve as force multipliers for our work. They don’t necessarily drive traffic to our site, but we influence how major finance stories are covered because we are often out faster than are with an informed point of view. And we have had some stories where we’ve been told that our coverage determined the prevailing narrative. That is what our readers pay for in our fundraiser, the knowledge that we can influence media discourse on our beat. Journalists overwhelmingly use RSS and Twitter. You don’t get click throughs on RSS unless you’ve got a clear (as in generally not short and cryptic) headline. Similarly, because I am on a nighttime production
  • 10. schedule (which you could have seen had you investigated) I have time only to auto-tweet headlines. Again, a more, not less, communicative headline has more impact. Meta Description.This discussion is unclear and therefore useless due to the failure to define what you mean by “pages”. If you mean pages as defined in the WordPress backstage, we do have six that lack meta descriptions, but they are all ones about the site (in our header bar) and while it would not be hard to add meta descriptions, it is inconceivable that someone conducting an external search would want to find those pages and be inclined to click through. So while this is not hard to do, this comes off as a “gothca” that adds to the length of your report but has virtually no real value added. If you mean a meta description for each post, that’s a completely unrealistic request given our resource constraints, which we told you about in our intake form. Separately, our WordPress support person tells us that Google made a major change of how it handled meta descriptions in January, with the result that it had a large negative impact on sites she works with that (unlike ours) are very concerned with keyword rankings. If you are going to tell people to provide better meta descriptions, and you are supposedly giving expert advice, we would expect you to tell us in much more detail as to how to create better meta descriptions, particularly since Google now appears to downgrade sites for repeated use of keywords/expressions in meta descriptions. Instead, you gave us three short and unclear paragraphs of apple pie and motherhood on this important topic, when this is a place where we need and expect technical detail, versus the four longer paragraphs on the irrelevant topic of testimonials. Outbound linking. This is the single most shocking, boneheaded comment in the entire report: “enormousblogroll”. You’ve just confirmed you’ve never
  • 11. done work for finance, economics, or political blogs. Thenumber of sites in my blogroll is low compared to norms for finance and economics blogs; blogrolls two to four times as long as mine are common. And political blogs typically have at least twice as many in their blogroll as I do (and we write more about politics than most finance blogs do, so the comparison is germane). You appear to have no understanding of the role a blogroll plays in the econoblogosphere. First, more than any single attribute, it tells new visitors who you are (as in what you read tells them what your focus is). I’ve frequently had readers tell me “You link to so and so often, you really need to add them to your blogroll”. Second, it cements alliances. Removing someone from a blogroll is tantamount to declaring war on them. Not showing it on every page would have a similar effect. It’s just not done. It would be like farting in a church. Whatever I might gain in Google points I would lose in the way of inbound traffic from aligned sites. This is absolutely essential to our site. But rather than offer a solution, you say, in effect, “drop the blogroll”. Your job is to solve problems. You need to give me a solution that allows us to keep the blogroll, not do the equivalent of throw up your hands. Site speed. Failure to do promised investigation. As indicated above, you appear to have missed that readers can access our site on small devices via opting out of Onswipe, which virtually all do. You apparently did not test the desktop version on a mobile device. More important, you failed to diagnose (which a cursory review would have identified) that the reason the site loads less than quickly is due to the ads and related tracking. That is served by and controlled by Investing Channel which had you looked in our backstage, you could have easily figured out. We’ve long been asking them to do something about it. As a result, your comments here and in 7.1 are irrelevant because you incorrectly assume we control these elements.
  • 12. Combine and minify CSS and JS files. Failure to investigate. As we’ve mentioned repeatedly, we run ads that come from a third-party server, so you didn’t identify what was under our control. We have only one CSS file. As for minifying, you missed several issues. First, we can’t minify the third party content. Second, how much you can minify is limited if you run more than 2 or 3 plugins and we have 18 active plugins. Our WordPress person, who has a lot of experience with these issues, has taken minification and combining as far as she can on our site. She says if she were inexperienced and had tried implementing your advice, it would break the site. Jetpack. This looks to have been a general cut and paste you dropped in and didn’t bother checking against how the site runs. Our WordPress person is very familiar with Jetpack, and uses it because it has unique functionality she values. She also manually (truly manually) goes over the settings ever time she updates it. But worse, this section has embarrassing statements that again confirm you didn’t properly review the site. “Since you don't have comments on most of your site„”. Naked Capitalism is known for having the best, most engaged, highest quality comment section of the entire econoblogosphere. If you had bothered looking in our backstage, you’d see we have over 450,000 comments. In fact, our comment frequency means we run the notoriously fragile WordPress database so hard that it poses a serious problem from a hosting perspective even with caching implemented. Similarly the “you are using several site stat packages” again reflects a failure to understand how the site operates. Almost all of the stat packages are for our advertisers (remember our discussion of Investing Channel?) and we can’t dispense with them. The
  • 13. WordPress stats are necessary in addition to that for our own use. Edit Flow. You didn’t bother understanding how we run this site and this is irrelevant and presumptuous. You see “multiple authors” and you assumed “hierarchical, scheduled process” which is incorrect. Had you looked, you would see that on virtually all days, one blogger is running the site. On the very rare days two people are DJing the site, we have clearly delineated duties (“you are doing X posts of Y type and I am doing the rest”). Activate Wordpress SEO/Check Wordpress SEO. We told you in our intake form that we are severely resource constrained. If you didn’t understand what that meant, the onus was on you to ask us. Among other things, it means that we are stretched so thin that we cannot take on any tasks that add to the time it takes for us to generate and publish posts (ie, the only changes we can afford to make from a time standpoint are at the site level, not at a post level). Similarly, if you had actually read any posts or comments, you would have seen that we are so time pressed that we often have typos in our posts. We don’t have time for a full copy edit. We certainly don’t have time to spend 3-5 minutes on SEO for each post. I am already on the verge of breaking down physically at my current load. In addition, there is a not-trivial possibility that the SEO will tell us Google prefers posts written at the 6th or 8th grade level to our reading level, which is college level. We aren’t dumbing down our content. XML Sitemaps. Another embarrassing proof of a failure to investigate. We do have a sitemap, called BWP GSX, which you could have easily seen in our backstage had you looked. * * * We have summarized our issues with your review in the accompanying Excel spreadsheet. Of your
  • 14. recommendations 1-25, we cannot implement most of them and the ones we can are ones you deemed comparatively unimportant. Specifically, if we throw out the recommendations that are: • not business-appropriate (1, 2, 3, 4, 11, 15, and the post- specific part of 19) • have no SEO justification and/or misguided due to failure to investigate (1, 2, 7, 8, 9, 10, 11, 12, 20, 25) • we don't have the capacity to do (3, 6, 13, 17, 18, 23, which makes 22 moot) • require significant content rework (6, 13, 17, 18) We are left with: 14, 16, 21, and 24, and a partial implementation of 19. Notice that none are ones you flagged as high- priorityrecommendations.From our perspective, there are three that could be: 4, 14, and 18. However, for 4, you didn’t give us a solution that fits our business imperatives, and on 14 and 18, you didn’t give us remotely enough guidance to be able to proceed. 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. So let me hazard giving you some advice. If you are offering a service that is customized, or you are pretending is customized, you need to qualify your prospects and you need to deliver a customized service. You did neither in my case, which is why the result was a fiasco.

Related Documents