Does CRO negatively impact SEO: Rivalry or friendship?
- Posted in Conversion Rate Optimization
It all started with a simple tweet from Elon Musk announcing that he acquired x.com.
Danny Sullivan responded with a light-hearted tweet about what ranks for x at this point.
Things would have stopped at this until Chris Portscheller injected himself into the conversation, dismissing the value of SEO when it comes to ranking.
It is surprising how some developers and UXers are still wrapping their heads around the benefits of SEO and the value it provides in building brands.
Equal to that is the lack of basic understanding of CRO and its principals most SEOs have.
I had the chance to speak at several digital marketing conferences in the last couple of months. And in each of these events, I made sure to attend at least one SEO audit sessions. These are sessions where attendees give their website URL and get quick feedback from SEOs on things they can fix on their site.
What is interesting is that in all the sessions I attended, 60 to 70% of the SEO feedback was not about SEO. The panel gave advice focused on usability and CRO. That is fine because it shows that SEOs are getting the value of CRO. However, most of the conversion advice given by SEOs was useless did not provide real value. The advice was on items such as CTAs, colors, and fonts. Something you might have heard a CRO talk about ten years ago.
CRO, SEO, development, and product must work together
Your online project (website, app, portal, etc.) will not succeed if those who are working in marketing, product, SEO, CRO, and development do not cooperate.
Each of these disciplines is a critical ingredient to your success. Without development, you will not have a product. And without SEO, people will not know about your product, and finally, without CRO, you will not have conversions.
Those who attack SEO and do not see its value miss a fundamental idea. One way or another you will have to pay to get visitors to your website. The most user-friendly website will NOT generate any revenue if it does not have visitors. If you use SEO to generate traffic, you will have to pay for SEO staff or an SEO agency. If you want to go the paid advertising route, you will pay Google or whatever other advertisers you will use to get visitors to your website.
Although I see some progress towards a better understanding of how these disciplines work, there is still some distrust that exists in between them.
Why do some developers and UX specialists think SEO is dead?
Every couple of years, you hear a wave of marketers/developers declaring that SEO is dead. These announcements typically come from people who do not work in the SEO industry. Mind you, I do not work in the SEO industry, but I have seen firsthand its power in both generating millions of dollars of revenue for business or completely crushing it.
Most SEOs either dismiss these claims that SEO is dead or they attribute them to the fact that the industry had few bad apples that gave it a bad reputation. But there are broader issues at work here. Ones that I do think SEOs will NOT be able to resolve because they are built into their industry.
1. We do not “fully” understand the Google algorithm
Coming from a development background, I struggled with the SEO initially. Not because I did not believe in the ability of an SEO expert achieve results. But most of the time, I saw them talking with such high confidence about the Google algorithm. I cringed when I heard a Google spokesperson speak about its algorithm, how it works and the different ranking factors.
It made little sense to me.
Can anyone honestly speak with high confidence about how an engineering team implemented an algorithm? Let’s say you are a developer and you write a 30,000 lines program on your own. You can speak with high confidence about the algorithm but can you guarantee how it operates in all instances?
There will always be instances where the code behaves in a way that you did not think of due to a bug or some missing logic.
Now imagine the Google ranking algorithm with tens of thousands of lines [I am not sure how big the algorithm is] and with many engineers working on it. Can one person speak with certainty about how the algorithm behaves in a particular way? We have seen this recently with some google spokespersons giving contradictory answers to each other.
Let alone the new complexities machine learning will introduce to the algorithm.
I am not saying that we should not trust SEO, but we should be careful about stating with certainty how Google will behave in a certain way.
How do you work around not knowing the Google algorithm?
Google is a commercial enterprise that has little interest in revealing how its algorithm works. This, of course, begs the question of how to understand its algorithm since success or failure online depends to a great deal on it. In this case, you have few options:
1. Follow whatever official guidelines Google publishes. I am not talking about what a spokesperson for Google says, but what Google officially releases as a standard. These standards are the primary rules to play the Google game. To a lesser degree of certainty, I would also carefully study and analyze what Google spokespersons mention. However, expert SEOs will check everything a Google spokespersons say against both the established Google guidelines and common knowledge in the industry.
2. Experiment. This is perhaps the most powerful way to figure out the Google algorithm. You change different elements on your webpage and watch how these changes the influence the SERPs. This approach is excellent in theory but difficult in practice.
- You must run your tests on a large number of properties to analyze the algorithms which of course takes up a lot of time and effort.
- There are ranking factors you do not control such as the behavior of other websites competing against you in the SERPs.
3. Analyze Google’s patents to understand where search is going. Understanding these patents can give you a clearer idea on how Google evaluates different elements on and off page.
2. Google keeps on changing the rules.
As Google changes the way it ranks websites, what might have worked in the past does not work anymore. And because Google has a [near] monopoly on search, it very much can do whatever it wants with little power to go around that.
There are many examples where Google
changed the rulesintroduced new rules or evolved its existing ones:
- After being a valuable element for an SEO strategy, meta-keywords have been entirely disregarded by Google for over a decade now.
- For the last couple of years, Google has been officially attempting to undervalue links, but different studies have shown that they still matter and have a significant effect on the ranking of your site.
- With the emergence of semantic search, Google uses a sophisticated approach to determine the topic of the site as opposed to keyword density.
3. Lousy SEO consultants gave the industry terrible reputation:
As the SEO industry boomed in early 2000, there was a large number of agencies and consultants who jumped in it as a way to make money online. Some of the bad reputation the industry suffers from is due to SEOs promising to guarantee specific results and not being able to deliver, or even worse, harming a website ranking and causing a Google penalty instead of helping.
The SEO industry view of CRO/UX
The SEO industry view of CRO has been a lot more accepting. After all, Google has been emphasizing the importance of usability as a ranking factor for the last few years.
If a person searches Google, the algorithm will display the best sites that match the searchers intent. When a searcher clicks and navigates to a website within the SERPs, Google hands over the job of keeping that searcher happy to that particular site. If the website provides good usability and high engagement, then the searcher is satisfied, and Google did its job. Yes, a website helps itself when a site offers right eyepiece to its users, but it is also ultimately helping Google.
For those who have been in the SEO/CRO industry for a while, we can all remember the advice from Google that SEOs, webmasters, and marketers should write for the human and not for the search bot. And if that happens, and users are happy, then your ranking will improve.
Nowadays, most SEOs are investing time to provide a better user experience and more conversions for their clients.
Revenue is what matters for an online business
How much traffic your website gets, how many social shares, and how much engagement it gets are all nice to have, but they do not matter if you cannot convert them into revenue.
Of course, if your business depends on the number of page views to make money, then you are all set. However, if you generate revenue by converting visitors into customers, subscribers or leads, then conversion rates are what matters to your website.
This is yet another reason why many SEOs are focusing on conversions. Most of their clients are asking questions beyond traffic or ranking. They are asking about the bottom line. How much money did you cost me this month, and how much money did you generate for me?
Evaluating SEO project from a CRO perspective
A poorly conducted SEO campaign can hurt conversion efforts. And most of the time clients are not aware of this. There are typically two ways we evaluate an SEO effort from a conversion perspective:
1. Evaluate the distribution of traffic across the website
A balanced SEO campaign should drive traffic to different areas of the site. I always worry when I see a single page or couple of pages on a website with more than 20 to 30% of the whole website organic traffic. Why does that impact conversions? Because if a particular page ranks for too many different terms, then visitors’ intent landing on that page will most likely vary. The page will fail to serve some of the visitors because they do not find what they are looking for and instead of continuing to navigate across the website, they leave.
To see the distribution of traffic landing on your website in Google analytics, navigate to BEHAVIOR > Site Content > Landing pages
You will see the landing page report. Notice the percentage distribution of traffic on your top landing pages. Are there pages that are getting more than 10 or 20% of the traffic? if the answer is yes, that might be a rad flag:
2. Evaluate how organic traffic converts compared to other channels
For well-optimized websites, organic traffic converts better compared to other acquisition sources. If you notice that your organic traffic performs worse compared to other mediums, then it could be one of two problems:
- you are ranking for incorrect terms or
- that your landing pages do not give searchers what they are looking for.
To evaluate this, navigate to ACQUISITION> All Traffic > Source and medium
A/B testing impact on SEO
I have written previously on how to ensure that your A/B testing does not hurt your SEO.
Here are few things that you want to keep in mind:
1. If you are using A/B testing software to manipulate a page(s) such as modifying the layout, replacing or add content – this should not cause any SEO issues.
Back in 2007, there was a concern that some SEOs can use AB testing for content cloaking, so their original page would be stuffed with keywords designed for search bots, while the page presented to users will have a better UX and better information. The bots can only see and index the original page which is designed specifically for them. The AB testing software is used to send 99.99% of the traffic to the well-designed page with good UX. With the sophistication of the Google bot, no one in his right mind would do this anymore.
2. Using split URL testing, where variations of a single page exist on a website at multiple URLs, could cause duplicate content issues.
This continues to be an issue especially if you are conducting a large scale split URL testing. For example, if you are running a split test across all the product or category pages on an e-commerce website where you have two challengers to the original design. In this case, every product page will have three different URLs: a URL for the original pages, one URL for the first variation and another URL for the second variation. You can see how this can cause a duplicate content issue. Unfortunately, there is no way around this.
There are few things that you can do to minimize its effect:
- 1. Use canonical tags on all the variation URLs pointing to the original page URL. The original page should not have the canonical tag pointing to itself. Remember that canonical tags are not a directive, meaning that Google can ignore them.
- Make sure that the header of the variation pages uses the no-index, follow direct to google:
<META NAME=”ROBOTS” CONTENT=”NOINDEX, FOLLOW”>
- Redirect (301) all the variation pages to the original page when you conclude your split test
- Make sure to run the test for enough time and not longer than needed.
3. Website loading problems due to AB testing.
Using an A/B testing software, there is a little bit of a delay when loading a webpage. And while this delay is minimal, it could add anywhere from 200 to 500 milliseconds to your webpage loading time. I have seen companies obsesses with trying to optimize their website loading time. But let me say this:
Going from 10-second to 4 seconds loading time will have an impact on your conversion rates. However, improving loading time from 4-second to 3-seconds will have less impact. At 3-4 second page loading time, the ROI gets less and less.
The conflict between SEO and CRO
We work closely with our clients internal and external SEO teams. And we have seen our fair share of conflict (and in one case, screaming matches) with them. Typical disagreements between the teams arise in one of the few instances:
1. The CRO team wants to change elements that can impact the ranking on a page
This typically starts with CROs suggesting changes to page titles or copy. In many cases, SEOs use titles and copy to send signals to Google about the content on the page. From a CRO perspective, we like to use the titles and copy to focus on the page value proposition and what is unique about the business. It has been my experience that both teams can come together to find the best copy that helps rank in Google and increase the website conversion rates as well.
2. Website navigation
This is a lot harder problem to solve. Some schools of SEO rely on website navigation (content silos) to send signals to Google on how content relate to each other. I have seen this technique work well with enterprise websites. The challenge is that CROs will also want to modify website navigation to increase the website conversion rates. While this can cause disagreements and heated discussion, it can be resolved through some creative thinking from the CRO side.
3. The SEO understanding of CRO industry
Many SEOs understanding of CRO misses the mark. Many SEOs talk about usability in terms of changing the color, size or location of CTAs, giving the impression that these changes will have a massive impact on the bottom line. Sometimes, SEOs give the idea that changing a word or a title on a page will also significantly impact conversion rates. But let’s be honest, CROs are the main reason for such bad advice. Many CROs do not use proper conversion optimization case studies or do not have a good understanding of AB testing statistics, sampling issues that could pollute their AB testing data, or even how long you should run an AB test for. You cannot blame others for your owns mistakes.
Should you focus on CRO, SEO or both?
We get inquiries on a daily basis from companies asking us to help with increasing their website conversion rates. The first question we ask ourselves is can we help these companies see a significant increase in conversion rates. To answer that question, we look at the website traffic and conversions to determine if the website is ready for CRO or not. Here are some simple rules we follow:
1- If your site gets less than 200 conversions per month, you are NOT ready for a full CRO program. Focus your effort on driving more visitors to your website using SEO, social or paid channels. This does not mean you should ignore usability and conversion best practices. It means that you will not be able to run a full A/B testing program because the website traffic does not support it. In our consulting (as with many well-established CRO firms), we require a minimum of 500 conversions per month (in addition to minimum revenue requirements).
2- If your website gets 200 to 500 conversions per month, then you are ready to start some AB testing on your site. You can typically run one to two test AB tests with one or two challengers to the original design.
3- If your website gets more than 500 conversions per month, then it is time for a full conversion program.
You should never stop investing in your SEO program or whatever channel you use to drive visitors to your website. The investment in CRO, however, is done gradually. When you have minimum conversions, you can look at basic CRO audits to make sure that your website works well. As you get more visitors to your site, you should move into a full CRO program.
Khalid Saleh is CEO and co-founder of Invesp. He is the co-author of Amazon.com bestselling book: “Conversion Optimization: The Art and Science of Converting Visitors into Customers.”
Khalid is an in-demand speaker who has presented at such industry events as SMX, SES, PubCon, Emetrics, ACCM and DMA, among others.
Join 25,000+ Marketing Professionals
If you enjoyed this post, please consider subscribing to the Invesp blog feed to have future articles delivered to your feed reader. or,receive weekly updates by email:
The Art and Science of Converting Prospects to Customers
By Khalid Saleh and Ayat Shukairy
- The Reason CRO is FLAWed
- No Low Hanging Fruits Here: Reflections on Running Conversion Projects
- Using Persuasive Web Design to Increase Conversion Rates
- Websites X-Rayed Series: A Deep Dive Evaluation of 200 Top E-commerce Cart Pages
- The Importance of Referral Marketing – Statistics and Trends [Infographic]
- Calculating Sample Size For An AB Test
- Speaker Diaries: Inbound 2018
- The State of Voice Shopping – Statistics and Trends [Infographic]
- Multiple Testing Problem: How Adding More Variations To Your AB Test Will Impact Your Results?
- AB vs. Usability Testing: Friends or Foes?