Directing Traffic and Thwarting SEO Misconceptions

Michael Martinez did digital marketing before digital marketing was cool. He is the president and co-founder of Reflective Dynamics. He blogs over at SEO Theory, where he delivers top-notch advice week after week.

He's widely known for questioning the SEO theories most of us take for granted. For example, he questions the value of many traditional link building approaches.

He's known for a deep analytical approach, too. Some of his posts are full of all kinds of arcane formulas and big-picture concepts, attempts to understand and explain the search engine ecosystem as a whole.

One of his specialties is helping websites recover from massive drops in traffic. If you're struggling with a penalty or an algorithm change, he's the man to talk to. If you want a fresh approach to links, content, and local SEO, or are struggling to repair traffic drops on your website or a client's website, read on.

Raney: You've observed a lot of problems with the way people are treating links as a ranking factor and as a source of trustworthiness. Can you tell our readers a little bit about what most businesses get wrong about links?

Michael: Everybody gets something wrong about links.

It's impossible for any web marketer, no matter how long they've been doing SEO, to get it 100% right.

Major search engines have multiple algorithms. They handle links differently on the basis of different things that are happening.

You might have one algorithm at Google that says: "Okay, we're going to go crawl this page," and you might have another algorithm that says, "We found this page for this link, but we're not going to trust it."

Or, "We'll trust it as a crawlable link, but we're not going to allow that link to pass value to whatever it's pointing to.

There are also links where people have used disavow files or nofollow link attributes to tell Google to ignore the link.

There's a lot of debate over whether Google follows those links or not. Google says they drop completely out of the link graph, but people have run experiments that indicate maybe Google is still doing something with that link.

So nobody in the web marketing industry knows what is happening with links. Therefore, everybody's getting something wrong. Just as a matter of principle.

Raney: What's a common thing that people get wrong?

Michael: Many people don't understand you need to think about more than one search engine.

In North America, you've got two search engines. You've got Google and Bing.

Bing is definitely the minority player here.

Outside of North America, you might be dealing with the Chinese search engine, Baidu. A third of the population speaks Chinese. Russia has Yandex. These are all just examples.

So, when you take on an international client you have to be sensitive to which algorithms they're trying to rank for.

Even in America, I've had people who have told me they just want to rank for Google, only for me to go in and do stuff that helps with Google but kills their Bing traffic. Which means they start losing money, and I have to go fix that. Each algorithm handles links differently, and you have to take that into consideration.

Raney: Given all of that, how should businesses address the linking process, if at all?

Michael: Let me tackle it from two different angles.

Let me take the search engine's POV first. Ideally what the major search engines who take links into consideration want us to do is ignore them.

They are trying to look at the web the way you and I look at it when we go out and click on the links and search for things. They want to try and index the web and identify the relationships between all the pages the way you and I see them.

They're not perfect. All the major search engines have flaws.

But that's what they're aiming for. They want to understand what's really important to people.

Take a query where nobody really knows the correct answer to a question. Search engines would still like to be able to show you content from websites that are more likely to be correct than not.

It would help them if we'd all stop trying to build links to influence their search results.

On the other side of the coin, you have site owners, either business owners or web marketers.

We're trying to build traffic to our websites from more than one channel. You have social media, advertising, offline advertising, you have the search engines, and you might have multiple websites. Big companies like Procter & Gamble have multiple websites and they'll sometimes interlink them to let people know they might have other products people would be interested in.

That's all natural. That's all acceptable. The search engines usually live with that. But somewhere along the way people start feeling competitive, and they start looking at all this data. And they say, "Search engines are using links, so if we get more links, that should help us."

In the long run, it generally does work. People will find new links pointing to their websites. They'll notice an increase in traffic at the same time. And they go, "Well, maybe I just need more links."

Over the past twenty, twenty five years, people have gradually pooled their knowledge based on experience. We've compared that to what we've been able to study. We've come up with this communal picture of "yes, links can help in several ways."

So, we all want more links, and are therefore trying to get more links.

That's where people get into trouble. Sooner or later, someone feels like their efforts to acquire links naturally are not keeping up with the Joneses. They start hiring web marketers to go out and build a lot of questionable links for them. That's where a lot of businesses have gotten into trouble over the years.

Raney: Can a business survive on earned links alone? Is there a better way for them to approach the earned link process that might allow them to avoid even the temptation of building the bad ones?

Michael: What would you say is the most successful website in the search engines?

Raney: Wikipedia.

Michael: They do not pay for links. They do not go out and request links. Wikipedia is a non-profit business, but it's still a business.

So can a business survive on the basis of earned links only? Yes. But not everybody is Wikipedia. Major brands like Procter & Gamble, Wells Fargo, and CNN have a lot of loyal customers and users. Those people will build the links for them.

These major companies still sometimes spend money on links. A few of them have gotten caught and penalized because they listened to (in my opinion) bad SEO advice.

I've always been reluctant to advise my business clients to go out and buy links, because they don't always understand there are penalties waiting for people who get caught. The more egregious your bad behavior is, the worse the penalties get.

For the most part, my partner and I try to earn our links.

Raney: How many links do business owners generally need?

Michael: Back when I had more time to do research, I found the vast majority of websites have few-to-no links.

So you really don't need that many links to get a moderate amount of search traffic. It's when people demand more search traffic that the demand for links starts to kick you.

Raney: You've talked about major brands and you've talked about international brands. Taking it a bit more local, how can a local SEO or local business establish authority in their community while avoiding questionable practices? Cause it can be hard from a content standpoint especially.

Michael: Let's talk a little bit of history.

Twenty years ago people depended on directories more than they depended on search engines. So, one of the first things I learned about in search engine marketing was you can describe your website more accurately or appropriately for a major web directory.

Of course, now they're all pretty much gone.

There are a few web directories out there and a few of them actually drive a little bit of traffic, but it's a trickle compared to what they used to do.

But we have alternatives now to those general purpose directories. We have business directories. People don't think of them as directories, but that's what they are. You have Yelp, Trip Advisor, GMB, Bing Places for Business.

These web directories are still very important to the small business community.

Some companies will hire web marketing specialists to get citations to help their GMB listings. Those citations are coming from other directories. If it's done well, businesses get traffic from those directories too.

This opens up a whole other can of worms, because once a business starts getting negative reviews, whether they're legitimate or not, they become concerned about their directory listings. And suddenly SEO doesn't seem like it's all it's cracked up to be.

Raney: Google's current strategy seems to be trying to keep people on Google's SERPs. How should businesses respond to that so that customers are still finding them?

Michael: While Google doesn't have an official opinion or response to that general opinion, they do try to discourage people from thinking that way.

They have admitted in testimony before Congress and the European commission that they're trying to solve the user's needs as quickly as possible. There's not a whole lot we can do about that from our side of the equation.

They are taking information and asking a question.

So some savvy marketers tell their customers: "If people are just looking to answers to questions, you're not going to lose any sales." If you're never going to get any benefit from that kind of traffic, the Q&A traffic, then what's the point in agonizing over it?

So, the real issue there for us as marketers is: can we make sure that the answers are properly attributed?

If a small business website is the source for an answer provided by a search engine—Bing and the international search engines are doing it too—what we want is proper attribution on these knowledge panels and featured snippets so consumers can see where the information is coming from, so it may lead to an eventual sale, or to some kind of relationship.

There are some industries where these types of questions lead to legitimate business.

Take a landscaping business. People will ask questions on the Internet like, "why are my plants sick," or "is my tree too close to my driveway?"

Those kinds of questions can lead to real business.

Raney: Another thing that a lot of people are getting very concerned about is that many of the strategies that used to serve them well turned into, or are rumored to be turning into, "pay to play" strategies. Are they right to be worried, or are there other things they should be focused on?

Michael: There's always been a really small percentage of things like that that turned into "pay to play."

Search engines saw an opportunity to make money from advertising and put advertising on the SERPs. Therefore, the advertisers got more traffic than organic people.

There have been a number of studies through the years that have held, fairly consistently, through the first 15 or 18 years when I followed this stuff, where if you had both high placement in the non-paid listings and you had advertising on that search results page, you got a lot of organic traffic. People weren't always clicking on your ads.

If people click on your ads and convert, then it becomes just a mathematical problem. It will pay for itself if it's done right. I don't think people should fear the need to advertise. What they should be afraid of is bad advertising.

Raney: You've been very successful at helping business owners recover from penalties and lost traffic. What's the first thing you do for them?

Michael: I look to see how badly they've shot themselves in the foot.

In a penalty situation, they have shot themselves in the foot. So you look at what the problem was, make sure it's fixed, and let the search engines know the problem got fixed. That usually gets the penalty lifted.

What's a lot more challenging are those times when people lose traffic to algorithm changes.

Raney: What's one of the bigger challenges when you're helping businesses with algorithm changes?

Michael: Sometimes what the web marketer recommends goes against what the site owner wants to do.

So we have to walk a fine line, because we don't know if we're right in our recommendations until someone follows through on them and we see the results.

The site owner doesn't know how to tell which recommendations are going to be helpful.

Thus, there's a point when you're helping someone with lost traffic in which you might both find yourselves in a less productive relationship.

What I often see happen is clients become frustrated and they stop doing business with the web marketing firm, because they don't know which suggestions are helpful or not. Which may be the right decision.

You might be dealing with the wrong bias on your provider side. You might need a new, fresh set of eyes to come in and look at what the situation is. They might suggest a completely different strategy that works better.

Raney: Can you give an example of a situation where your recommendation went against what the business owner wants to do?

Michael: A couple of years ago a news website came to us. We did an audit for them.

They got most of their traffic from Facebook. They had been doing that for years. And they wanted to build traffic from Google. They felt there was a market for the information they covered. They wanted to hedge their bets. They weren't really sure where Facebook was going to go. That was a good instinct on their part.

So, they were actively building up their Google traffic. And a few months before they contracted with us, they had a problem with Google. Suddenly traffic went down.

We looked at a number of things they had done. They had made changes to their website between the time they lost traffic and the time that they brought us in. And that always clouds the issue for the SEO doing the audit.

But, I noticed they had a video, a single video, that was being displayed at the top of every one of their pages.

Based on other technical things that went into my analysis, I recommended they either remove the video or encapsulate it in an iframe so the search engines would be less likely to see it.

They didn't want to do that. They wanted the video to be their brand message.

If the web marketer makes a recommendation that the client doesn't even attempt, then the web marketer is in a very bad position. Because we can't hold ourselves accountable to our own strategies if we don't see the feedback from trying these things.

The website is doing fine now, but because they didn't follow through on all the advice I gave them, I don't know how much my strategy would have helped them. I don't know if they're doing fine because they benefited from something else, like a random change in the algorithm.

These types of situations happen to web marketers all the time. People get very frustrated, because it's not an exact science. It can be a very painful process for both the web marketer and the client. The client is often following a sunk cost fallacy. "I've lost 80% of my traffic. I'm afraid I'm going to lose the other 20%."

That's a very powerful emotion to go up against.

Raney: What do SEOs or local SEOs tend to get wrong when they try to help their clients?

Michael: Expertise, authority, and trust. EAT, as it is now well known. It's a concept that comes out of Google's Quality Rater Guidelines. Many people have expressed frustration with the proposed strategies for improving EAT because it doesn't work. And the reason why is they're trying to game a metric that doesn't exist.

Google tests their algorithms all the time. And part of that testing process involves showing search results to third party contractors called Quality Raters. Normally what they see are two screens. So they'll be given a query to run with two sets of results. One set might be a live search result. The other set might be a test search result. The methodology is generally reliable.

What Google tells the Quality Raters to do is to look at the sites they are shown, and rate them for the test. These ratings don't impact search results rankings in the live system. But what they do is they provide feedback to Google's engineers about how good each set of search results is.

They are asked to assess things like expertise, authoritativeness, and trustworthiness for these websites.

It's a human evaluation. There's no algorithm that does this.

It's an indirect assessment on Google's part. And unfortunately a lot of people now are going out and telling business owners that they need to build up their expertise, authoritativeness, and trustworthiness. Which in a human sense is an admirable and desirable goal for every business.

But if you're trying to do it for SEO you're just shooting in the dark. Some people get luckier than others when they do this.

Raney: What types of content strategies are you steering your clients towards these days?

Michael: It really depends on the market.

I always recommended people look at what their customers actually need and address those needs.

What people did for a long time was they looked at keywords that were generating traffic, and they just wrote copy around those keywords.

It didn't really matter to them what the copy said as long as they got the traffic.

For most people, that's an inefficient way to market a business.

You can't use that cheap copy the way you used to. It still works a little bit, but you need better written content that specifically identifies a need, proposes a solution, and offers a rationale to why it's a good solution for that need.

The more you do that, the better you tend to do overall.