Ask HN: Does SEO still work in 2018?
If it works, which method are you using now?
Instead, write 'honest' content. Write it well. Write it for humans. If you find yourself adding content in hopes the search engine will pick it up, that is a short-term gain for long-term pain. Do have your title match your content. Do try to write unique phrases that are applicable to your content. And try not to copy/paste content around your pages.
Do all that, with clean markup, and your SEO will be good enough to beat out local competitors. (Even if agencies tell you differently.)
The one thing that is timeless, that works, is good content.
If you write good content you will get far.
Flooding pages with keywords and link farms doesn't get you anywhere in the long run.
This is not true. Google has been gamed for well over a decade.
On a past enterprise, I believed the lie "write good content and it will magically rank" and paid a content writer for about 9 months. She posted lengthy original content about 3 times per week, all highly relevant to the search terms we cared about. I did the same about once a week.
A competitor with a private blog network and professional spam apparatus started outranking us just with "link building", i.e., posting links on the PBN they controlled, within 1 month of launching (we had about 15 months of content at this point). They had no content on their site at all, just a landing page that included a few of the keywords they were using in their link building. There was no blog, there was no "well-written, honest content for humans".
Google had no qualms ranking them based on the link text despite their lack of supporting content. This competitor brutally beat us in organic search traffic until both businesses were shut down due to legal threats from a F500 cowering behind the CFAA.
Our blog is still online. There is no significant search traffic to speak of. While the content may have aged some, it's still the best resource for a lot of the things we posted about. It's buried pretty deep on Google's search.
Transparent spam and plagiarized content are easy for Google to detect, but private blog networks are a major component of modern SEO, as are several other non-organic schemes that Google has no simple algorithmic method to detect (barring slip-ups that make it obvious, though even in those cases, Google seems reluctant to punish domestic PBNs). I know of entire offices staffed with "link builders" whose entire job is to figure out how to get backlinks, and to be fair, "write really good original content" is somewhere on their list, but it's not very close to the top.
Just a counterpoint. "SEO" is still a thing. Google is a high-value target and it's worth a lot of money to game them. It is quite naive to believe that they're impervious to it. You can rank OK with "honest, well-written content for humans" as long as no professionals are competing with you.
Agreed. Google is certainly advanced, but it certainly isn't fair, despite what their guidelines tell content writers and webmasters.
Google wants people to write high-quality, relevant content, so they altered their guidelines to simply tell everyone to write high-quality, relevant content. Yet, just as the YouTube lottery exists -- where two creators can produce videos of similar quality and one sees enormous success and the other doesn't -- there's probably a ranking lottery.
And at the end of the day, if a content moderator or employee at Google likes your content, you'll have a better chance at ranking regardless of quality.
Similar to startups, I am curious as to how much of professional SEO success is about outreach. i.e. if I am mates with a uni professor and I write something about his topic, can I get a backlink from their valued .edu domain? What about .gov? Does this outrank better content? To what extent are forbes.com, entrepreneur.com, inc.com etc writers gaming this? (Or are they just looking for the virality?)
Is paying a content writer (who's area of expertise is, usually, being a content writer) a more noble attempt at gaming the system, I wonder?
But, I think it's a valid question since a lot of paid for content borders on being nothing more than an SEO game plan?
Paid content writers will often produce work that's well-structured, easy to read and grammatically correct but is just a rehash of something that's already been written about 100 times before.
I once employed a journalist to do a piece about a particular kind of insurance. She did a fair bit of research and even got quotes from prominent people in the industry.
I've seen the same piece many times in various guises, all well written but, essentially, her article with just enough of the words changed to make it 'unique'. Easy to find too since they use the quotes verbatim.
When challenged the response is always the same - 'we paid someone in good faith...'
For most people, they run a business, and their web presence is a tool to improve that business, and SEO is a tool to improve its effectiveness.
But cookiecaper seems to be talking about a totally different kind of SEO, in which the base goal is to generate traffic and monetize it, so SEO becomes the primary tool by which you look for easy keywords to target, and write content specifically for the search engines, not for a human audience supporting a pre-existing business.
That entire scenario is fundamentally based on gaming SEO. It is a short-term play for cash, and some people do well with it. But we're comparing apples to oranges when trying to match that up with the way an actual business would approach SEO.
Most of the aggressive SEO players I know about are supporting "actual businesses", not content farms that survive on ads. Our aggressive competitor had no content to monetize directly and no ads, they were competing with our service.
Note well that suggesting Google may not be all-knowing draws out the lackeys accusing anyone who may speak this heresy of not running an "actual business". Implying, yet again, that the only reason someone would speak ill of Google is because they're too perfect. Give me a break.
All of this misses the point regardless, because it's not "gaming" the system when you're playing by the rules. Google wants you to believe that its algorithm is perfect, so it says the best way to rank is to write good content, which they will magically identify and publicize. So you pay someone to write good content for your site and, as long as this content is actually original and useful, you're doing exactly as Google directs. That's not what "gaming" is.
SEO is still a thing because people come in naive enough to believe that good content will stand on its own merits, like I did, and then quickly learn it doesn't actually play out that way in real life.
I've seen many businesses including my own boom with good SEO. Back in the old days I got to #1 for "movers", but the tricks back then don't work anymore (which is a good thing).
Today, it's much more legitimate and I find the most success with keyword research. It's about finding what your target market is searching for and where your competition is or isn't.
Giving up on SEO is a losing proposition for any business that doesn't have a dominate brand or social media presence, which is most businesses.
For site maps, it helps Google find updated content if you have a large site. It's not required, but it won't hurt if you submit one.
Disclosure: work for Google nowadays but have zero insight into anything internal that's search related. The above is just common sense and should apply to most search engines.
From my experience if you're using any number of JS frameworks, this is almost impossible. If you're building a static site, you're better off just writing flat HTML or using a well known framework like Bootstrap or Foundation.
Reading a lot about it and experimenting with many sites is the best way to learn. It requires more than just good content if the keywords are competitive.
I joined a startup in 2010 when SEO was all the rage. I was the second tech hire, and my boss left pretty soon after I joined because he was sick of managing and wanted to be a dev, so I became lead dev with another guy under me.
The tech was still relatively new to me, but I had previous knowledge of SEO from a few internships I had done while at uni. I also freelanced on the side, doing SEO audits for a few local companies. We ran a service in the legal space, offering leads to law firms from customers interested in a particular type of law, so the name of the game was to get as many eyes on the site as possible.
The MD wanted to follow a similar technique to our parent company, which was working well for them. This was essentially paid-for spam, and I felt really uneasy with this approach, so I suggested that we ignore a lot of the current SEO tactics and respect Google for what it is - a very smart search engine built by people that are smart enough to fix the common SEO tactics of the day. Thankfully, the MD listened to the dev team, and the focus was on building the best service possible. Instead of getting crappy backlnks, we ran competitor analysis tools to figure out why some people were doing well, and making our service better than theirs. Essentially, instead of playing the game, we wanted to deserve the top spot. We had content writers reviewing content regularly, a marketing manager that (eventually) backed us, and a manager that was happy to have a wildly inexperienced dev team fuck things up in the name of getting better.
Things went well for us, and we made some slow progress. After a while the Panda update happened, and we skyrocketed to the top of a lot of valuable terms, while the parent company (using spammy technqiues) crumbled, losing around 70% of their traffic. I think investment bailed, and we split out on our own. We ended up getting millions of visitors a month, and the company was acquired because of the work that was put into the product.
Since that experience I've almost completely shunned the SEO side of things, favouring organic online marketing. I know that a lot of the techniques that agencies would use worked - I saw them work myself on our competitors, but I also knew that they were a short-term fix with no long-term gain. By focusing on clean markup, good content, a clear architecture of information, and fast loading times, we beat the competition, and those values haven't changed eight years later.
The company handled data on investing, but their CMS was full of pages relating to totally unrelated things like Nokia phone repairs. It was the standard long tail approach, where a thousand pages of spam getting 10 visitors a day was enough to give them a legit boost on their main pages.
When the Panda update hit, these pages killed the site, and the spammy links propping these pages up were next to worthless.
I get a lot of search engine traffic to my blog, with the following strategy:
Stage 1: Technical - Clean up your act.
If you structure your site properly and make it accessible, search engines will crawl and index it properly. Make your content accessible from a technical standpoint.
Clean up your HTML
Make your URLs reflect your content, cleanly
Build and submit sitemaps
Make the navigation on your website simple and effective
Make your website load quickly
Stage 2: Content - Make great stuff.
If you build great content, people will share it. They will link to you.
Build great content
Make things that help people
Stay within a niche if you can
Create content that helps people, even if you're giving a little of your service away. (For example, a roofing company featuring an article about how to choose a roof, even if it sends them to your competitor, or some DIY stuff)
If you do these things, you'll do well in search engines. I'm oversimplifying a bit, but this is way more effective than keyword stuffing, link purchasing, etc.
Just keep it honest.
As a side note, I noticed that simple quick-loading pages seem to have more of an advantage recently than people think.
If you're an ecommerce shop optimizing product and category pages - SEO will help.
If you're a plumber and ranking higher than your competition will bring in new leads & jobs - SEO will help.
If you're an established SaaS company launching a new product - SEO will help.
If you're a brand new SaaS startup - SEO is probably not the best thing to spend money on. That money is likely better spent on PR getting press mentions, and on PPC ads promoting your brand.
The main problem I see many small businesses making is relying solely on SEO to bring them leads while having no other marketing strategy. This may have worked well for them back when you could spam websites to the top of SERPs, but it's not the case today.
High reputation sites: travel, hotels, the popular marketplaces.
Junk: drugs, mattresses, pills to grow your hair back, cooking recipes, SEO.
In my experience, it's brand new or still relatively young companies that SEO is often not great for. It might be a great thing to engage in in the future for them, but early stage companies strapped for resources need to be more focused on building out an actual brand rather than on ranking page 1 against competitors who may be spending 10-100x on marketing than them and have been around for years.
1. Everyone knows who the competitors are 2. You are either going to be paying for those services or not, and if you are you will decide which competitor you want to pay for based on something different than how they show up in your google.
Examples - Thomson Reuters' legal services, LexisNexis etc. I'm pretty sure the amount of times some lawyer searches for a term, ends up on some competing legal service and just says oh well I better just pay for this now is negligible.
Near or actual monopolies.
SEO in these cases might help rank, but you don't actually need to rank for your business.
I've also been starting to notice that having custom HTML / CSS is boosting things more than before for smaller sites.
Also, even if using JS for SPAs doesn't matter to Google there is a reasonable long tail of other web crawlers that may or may not be helpful to your ends and you'd be surprised at how many of them just use mechanize or nokogiri instead of something more robust. Basically it comes down to cost, last time I was in the custom crawler trade it was around 50x more expensive to do things with a virtual dom than it was to either straight parse the HTML or figure out the main API calls JS was making from the browser that actually had the data I needed and just run those.
Selenium / headless Firefox with JS running in the browser waiting for things to load / resolve. This is the worst case scenario.
Sometimes you can use a layered approach, first do a normal mechanize request then if the page doesn't 404 you can try to parse it there. If you detect that its relying on infinite scroll try to run the JS command via Ruby Racer then if that isn't [semi-]automatically possible put the URL in the "requires full DOM queue" all of this on a bunch of code that lights up different servers to evade log analysis. Some URLs are behind a CAPTCHA, and some people will then add those urls to the "Use a human to do the work" queue, but I've never really felt right about doing that. To me a CAPTCHA is basically someone saying, "no seriously, bugger off" plus I don't want have some poor soul waste their life copying and pasting text. It's easier to just convince a client that last 0.01% of data isn't worth getting.
What's really funny is the meta game that stores have. There is no "price" any more for many of these people. There is "the price you happen to see" and they use a bunch of tricks to try to figure out if the traffic is real or not because they know that other places are dynamically setting their price to undercut them by some optimal amount.
It's just an arms race, but you know what arms races are really good for?
Arms manufacturers.
If you want to be featured at the top of search results you need to have content that is actually relevant to the search term. It has to be genuinely relevant and of high quality and ideally, linked to by other respected sites.
You can spend all the time and money you want on the gimmicky stuff, google will update the algorithm and wash away any progress you make.
/s
Seems to me like it's a decent resource nowadays.
https://web.archive.org/web/20110117085131/http://w3fools.co...
Even if w3schools have fixed all these instances since then, that all those errors were ever produced is a bad sign, especially when there were better alternatives even then.
https://chrome.google.com/webstore/detail/remove-w3schools/g...
This is the key. Google wants to make sure your readers are getting their questions answered. SEO is constantly changing, but high quality and valuable content will always be king.
It's an association engine.
We know what Google is trying to do - associate the best solutions to search questions.
To game Google you need to help them achieve their objective - associate their search queries with your solutions.
Broadly there are three types of search desire:
1. navigational - the user wants to find a resource or place?
2. informational - the user wants to learn something?
3. transactional - the user wants to perform an action (pay a bill, book a flight, hotel, restaurant etc)
If a page satisfies a user's search desire - they won't search for it again.
This missing search trail is a valuable signal to Google. It provides some evidence that page A satisfies query B. The more these signals fire the stronger the association.
SEO 2018?
Ideally make pages that solve your users' search problems without them needing to go back to Google to search again.
As long as google uses a metric to determine how good a page A satisfies query B there always will be the possibilty to optimize for this metric instead of actually to solve users search problems.
That said, if you're targeting a certain city, definitely still works. Most local sites are poorly optimized.
The changes Google will do to their algorithm will sooner or later be gamed, SEO 'experts' will beat a small subset of the system, Google will penalize them and adapt making it vulnerable to more exploitation.
The search engine is interacted with by humans so no matter what Google does, at the end humans(SEOs) on the other end will try to present their content first the way they see fit for humans.
I found out a number of things when I was working closely with my clients. Most were doing ok in the SERPS and a few did really well because of the niche industries they were in.
Some things I found out:
1 - SEO helps, but is never the determining factor in hiring any of the companies I worked with. This means you can be #1, #1 in Google and still not get any business because of it.
2 - Tracking contacts is critical. We mandated all of our clients use an answering service. They would contact people within 24 hours of receiving a contact form. The owner would then follow up to set up appointments, consultations, etc. Once we did this, we saw a massive jump in sales conversions which had nothing to do with SEO.
3 - If your design sucks and people can't find what they're looking for quickly, they're going to leave your site no matter how high you're ranking. We found this out with several of our clients who didn't want to redesign their site for various reasons. They ranked high, but their traffic and conversions sucked. We redesigned their site(s) for easier navigation, and prioritized content and got rid of other content that clogged their site. Conversions jumped by 43% over a 6 month period.
SEO does matter, but even if with massive amounts of traffic because you're ranking well isn't a magic bullet for success. You also need a good design and content people can find quickly and easily.
From what I see, in the UK at least, small businesses aren't doing this. Either they can't be bothered or are afraid of giving away knowledge they think they could be charging for?
Plugging my own project but based on the above principles I wrote a guide containing on-page SEO rules that authoritative sources say are important (e.g. Google) and created a Chrome extension that tests your site follows these rules:
Edit: Let me offer a different way to look at it.
Does having a well organized website help turn visitors into clients? Yes. Well that's SEO, a well organized website.
Does having useful and well written information about the services and products you offer, or you industry, help your business? Yes. Well that's SEO too.
Can sharing your knowledge online across different relevant forums help you grow your brand and business? Yes. Well that's SEO as well.
It is just that people who do not know SEO think it is things such as keywords and metatags and url structure, but everything above and more is SEO. And all these things work to your advantage, therefore SEO works.
SEO is often more jumping between things that don't fit most peoples preconceptions of SEO than it is about editing a title tag or submitting a sitemap.
The problem with SEO and why I have a grudge against the industry is past success by gaming google with PBNs has led to hoisting up a few sites that can dominate because of their past. Those sites used spam tactics, then ranked high, that led to actual relevant backlinks, google got confused, and now keeps them at the top for any new content. Others must also utilize these dicey methods and pour serious money into it to do well. At least nationwide. I'm praying for the day that house of cards comes tumbling down.
Until then, having a better site, with better copy (done by a professional writer not farmed out), and using Adwords typically is the go-to.
I haven't touched Family Law but it might be similar to PI. PI on Adwords is brutally expensive and that's why hyperlocal SEO works well in that regard.
Currently, most Japanese Google search results are garbages. It's all boilarplate texts(but produced by very cheap human labor writers so you can't easily filter it out) with keywords in it.
SEO won, Google lost.
After that comes paying for AdWords to get those reviews in front of more eyeballs.
Similarly, AMP has kind of changed the game in terms of publisher traffic. Users seem to be assuming that results that do not appear in those carousels are more outdated / less relevant.
http://plumber-plumbers.co.za/index-rivonia-plumbers.html
http://www.plumbersgauteng.co.za/gauteng-plumbers-areas/plum...
That is so gamed. Keyword stuffing, repeated exact pages. I assume it's better in the US but surely the algos are run over the whole dataset? Is this a worldwide problem?
When I first did that search I found nests of sites using exactly the same content and styles, with different areas. Same phone number across many sites, cross-linked out the wazoo. I don't have the interest to check now but it doesn't seem much better.
If something of that profile can’t get ahead of pages gaming pagerank, independent blogs focusing on content have zero chance without at least doing some legwork on backlinking and keyword optimization.
High quality content isn't cheap and you have to be in it for the long game.
Everything Calories Per Dollar bring in hundreds of people to my website a day. As a result, I studied more restaurants Calories Per Dollar, studied Calcium Per Dollar, and other things like Caffeine Per Dollar.
All relevant to my website, all quality content IMO. Google is easily my best source of traffic.
-t. my impression from dabbling briefly with it.
But more often than not, SEO is an umbrella term which encapsulates both the good and the bad practices. Given the work Google and others have done on weeding out the latter using ML/AI as well as human intervention [0] I would say you can't be sure whether those would work. It's contingent on Search Engines being aware of the practice and penalizing that behavior. So it may work today but not guaranteed to work tomorrow.
This saves the user time because they may get what they need directly from the intelligent search engine result data so that they don't have to enter my site. On the other hand Google makes deep links into the site immediately available as "sub-results" so that a user can enter the site more directly without wasting time on an entry point.
Really, SEO means labelling your content so google understands what it is.
After that, it's all about links.
The question itself begs for an anecdotal answer that is completely irrelevant to your circumstances...
More importantly, the amount of people who have given up on strategies too soon or because it wasnt done in a way to provide empirical evidence, is staggering... and the amount of misinformation that flows out into the wild built on those flawed premises is really problematic for those without the context of your situation and the relative context of the situation where those opinions were formed.
The second problem is that opportunity cost is rarely measured by marketers. The third problem is that your resource allocation relative to your objectives is arguably more important than any other factor or opinion about a strategy.
Marketing decisions should be made in the context of your resources, goals, and your broader marketing mix.
The idea that you can evaluate a specific strategy and measure its return in a Silo without considering the opportunity costs and other available strategies relative to your specific resource limitations is very dangerous and needs to be killed off!
The ideal scenario is to optimize to cashflow relative to lifetime value and time value of money...or get as close as possible to this...and choose strategies within the context of this calculation.
You want an SEO strategy? First, figure out if your audience is there and the cost to reach them with paid search and how long it takes to turn into cashflow and the lifetime value of that customer...than evaluate the costs to optimize to rank for that keyword, and how long it will take, than look at your other resources and strategies relative to the likely costs and returns and time to get cash returned relative to those strategies and decide if its worth investing in ranking there...
Once you identify a proven keyword and have context for the cost and value and time and oppirtunity costs, than figure out how to rank for that keyword... maybe you buy a site that is already ranking. Maybe you advertise on sites already ranking. Maybe you think about the user and find a way to create a resource that Google will want to rank because of its utility... and than create empirical tests to measure progress and than you invest !!!
------ The shortcut SEO strategy i like is to ask yourself, what resource can I create for my industry that is not online but would offer so much utility that google has no choice but to optimize their algorithms to rank my resource... like wayne gretzky, go where the puck will be in the future...
When is SEO is done right, its Google trying to optimize its algorithms to rank your content, not vice versa!
Just my two cents.
The guy Brian Dean has many years of experience about SEO
The answer is SEO matters a lot. Yet SEO today means something entirely different from what it meant a few years back.
Back in 2015 when I started my blog, I was writing what would be defined quality content. It was very in-depth, researched (at times it would take me weeks to write a piece of over three thousand words). However, when I hit the publish button I wasn't getting any traffic (at the time a few dozens visits per month!). True, quality content is the baseline, but if you don't have an idea of how search engines work, you might gain traction over a few years time, but the process is quite long and not guaranteed to succeed!
However, what SEO means today is different from what it meant a few years back. In the past keywords and backlinks were all that mattered. They still count. However, it will come a day when they will stop working. Why?
Since 2012, Google has been building a giant knowledge graph, which main aim is to gather information on the web, structure it in the form of nodes and edges to construct this massive graph. Thus, getting featured within Google's so-called knowledge vault becomes critical for an effective SEO strategy.
Thus, if in the past you might be targeting keywords to be positioned on the SERP. Today you might want to be more focused on long-tail keywords that can bring you in featured snippets or inside a knowledge panel. In fact, those are also the avenues to get into the Google assistant, thus voice search.
In short, while in the past it made sense to use backlinks and keywords because Google lacked the power to use more sophisticated methods to index and rank the web pages. Now you have to switch from a keyword centered content model. To an entity-based content model.
This kind of approach is more holistic and targets not only the SERP, but it allows you to get ready for voice search. What are some practical examples of having this approach?
- an internal glossary that targets specific questions your readers/potential customers might have - structured data in the form of Schema injected as JSON-LD (in this way you're giving better data that does not affect the performance of the page) - use a better internal linking strategy, that connects articles to topical pages (like the glossary pages) and back
There are more strategies to adopt, but those are the ones that come to mind right now.