Every search engine optimization (SEO) issue, from a sudden drop in organic search traffic to a decrease in conversions, can be traced back to one simple truth: Something changed.
It’s tempting to panic at first, but a shift in perspective is crucial to keeping up with changes in the competitive landscape and the ever-evolving search engine algorithms. Instead of viewing these dramatic changes as catastrophic events, we should be treating them as solvable mysteries.
Gone are the days when reactive, “best practice SEO” is the only system in place.
Over the years, search engines like Google and Bing have been adapting to the ways people search for content, putting more emphasis on concepts like user intent and the topical authority associated with a website. In order to keep up and create long-term proactive fixes, you must adopt a data-first approach.
I like to think of data-first SEO as one giant scientific experiment. Many of us learned the scientific method in school:
Ask a question > Complete background research > Create a hypothesis > Test your hypothesis through experimentation > Analyze your data and draw conclusions
That same framework that drives scientific discovery can be applied to SEO.
In this guide, you’ll learn how to use data to solve any SEO problem using a 7-step system that checks for troubleshooting opportunities from content strategy to technical SEO.
What is Data-Driven SEO (and Why It’s Better than “Best Practice SEO”)?
Data-driven SEO uses metrics and data sets to inform every step of your content strategy, optimization efforts, and technical SEO. Analytics takes some of the guesswork out of SEO, allowing you to focus on what works and what doesn’t with the empirical evidence to back it up. For example, engagement metrics like bounce and conversion rates can provide insight into what content resonates with your readers, and analyzing technical data like indexation reporting can unearth issues with crawling and content quality.
On the other hand, the standard “best practice SEO” model relies heavily on a universal checklist based on generalizations about how search engines have worked in the past. With so many factors constantly fluctuating in organic search, adopting a data-driven approach enables you to strategize more effectively and target long-term benefits.
Essentially, data transforms SEO from opinion-based decisions into an evidence-based strategy.
7 Steps for Using Data to Solve Any SEO Problem
1. Clearly Define the Problem
Instead of making a general statement like “Traffic is down,” be as specific as possible, including timeframe and impact, to determine the priority:
- Organic search clicks to URLs containing the URL fragment “/blog/” decreased by 32% over the past 3 months;
- Organic search conversion rates last month changed 0% despite a 15% increase in organic search
- Organic search click loss for the last 28 days, year over year, has been isolated to keywords containing “best.”
When you have a clear problem as your starting point, you can build a more targeted plan.
2. Gather the Right Data
Once the problem is defined, the next step is to do your research and collect the necessary data. Creating a comprehensive plan of action works best when you cross-check data sources to pinpoint the root cause of your SEO problem.
The challenge isn’t finding the data. We have access to more information than ever before. But you can’t solve every SEO problem with every tool. The real challenge is knowing which metrics matter and what decisions can be made from them, because each datapoint in each platform measures something differently, and each serves a unique purpose.
Here are some common SEO tools and when to use them:
- Google Search Console: Use for visibility problems. Check keyword rankings, impressions, clicks, and clickthrough rates, and identify technical issues that can be hurting your visibility on the search engine results page.
- Google Analytics 4: Use for user behavior problems. Compare performance across digital channels and identify issues with conversions and other user behavior.
- SEO platforms: Use for opportunity problems. Analyze competitive data and content gaps.
- Crawling tools: Use for technical problems. Identify issues with crawling, tagging, duplicate content, error codes, and more.
- Backlink tools: Use for authority problems. Analyze the sites linking to your site and your competitors’ sites.
3. Segment Your Data and Look for Patterns
General averages and percentages can often hide important details. This is where data segmentation comes into play. Segmentation acts as a tool to pull back the veil and expose the issues in your SEO program.
Some common ways to segment your SEO data include:
- By page type: Analyzing performance based on the type of page (blog posts, product pages, etc.) helps determine if the problem is related to pages using a specific template or containing a specific type or intent of content.
- By keyword query: Segmenting your data based on specific search terms that drive performance can allow you to narrow your focus to specific content topics and competitors. Analyzing branded keyword performance compared to non-branded performance can be especially helpful, especially in combination with other topical themes.
Once you’ve segmented your data, you can begin to look for the patterns. For example, you notice a decline in traffic only on non-branded keyword clicks to your product pages while clicks on branded keywords remain stable.
4. Diagnose the Problem and Form a Hypothesis
With a focus on specific data pertaining to a specific issue, it’s easier to get to the root cause of your SEO problems and determine a targeted action plan to fix it.
Example
- Observation: Organic traffic to blog posts that received clicks for queries containing the word “reviews” dropped 32% in the last 90 days. Previously, the affected blog posts received 1,000 clicks per day on average, and now they are receiving 780 clicks per day. After checking the competitive data, I found that rankings for Competitor A have improved, perhaps due to recent content updates.
- Hypothesis: If we rewrite outdated content and address content gaps, then the topical relevance and freshness of the affected blog posts will improve, causing rankings to improve and increasing traffic to the pages.
- Actions:
- Refresh the content in the title tag, meta description, headings, and body content for affected blog posts, adding content on recent industry or topic updates, and aligning with the keyword themes’ current intent.
- Add a new section to relevant blog posts that features a case study to strengthen E-E-A-T.
- Add an FAQ section to each affected blog post addressing related questions found in Google’s People Also Ask section, customer service call topic data, and keyword research.
- Expected Outcome: Ninety days after the refreshed blog posts go live, the number of clicks per day for the affected blog posts in Google Search Console will have returned to 1,000 clicks per day on average.
5. Implement Changes
Now that you’ve laid all of the groundwork, remember not to post all of your heavy-hitting changes to the site at once; otherwise, how will you know which part of your plan actually worked?
6. Track Changes and Measure Impact
The final step to your data-driven strategy is to test the performance of your action plan to determine if it met the expected outcome.
Don’t forget to give Google enough time to crawl and reindex your pages, especially for the more involved updates. You can speed this process up by requesting indexation for the updated pages in Google Search Console’s Inspect URL tool.
7. Repeat
Once you are satisfied with your findings, refine your strategy. SEO is not a one-time fix. As the way users search evolves, search engines will continue to adapt. However, with a framework deeply rooted in data analysis, SEO mysteries become more manageable, especially as you develop a repeatable system you can use to solve virtually any problem.
