SEO Journal: How to Track What’s Working

An SEO journal is a structured log of your search optimisation activity, ranking changes, and the reasoning behind every decision you make. It turns SEO from a series of disconnected tasks into a traceable body of work where cause and effect become visible over time.

Most SEO practitioners track outputs: rankings, traffic, clicks. Fewer track decisions. An SEO journal closes that gap, giving you something most teams never have: a reliable record of what you tried, why you tried it, and what actually happened.

Why Most SEO Teams Have No Idea What’s Working

When I was running iProspect, we managed SEO programmes across dozens of clients simultaneously. Some performed well. Some underperformed. And when a client asked why their rankings had improved or dropped, the honest answer, more often than I would have liked, was that we were working backwards from the outcome to find a plausible explanation.

That is not a confession unique to one agency. It is the default state of most SEO work. The feedback loop in search is slow, the variables are numerous, and Google does not tell you which change moved the needle. So teams fill the gap with post-hoc rationalisation dressed up as analysis.

The consequence is that good decisions and bad decisions look identical on a spreadsheet. You make a change. Rankings shift three weeks later. You attribute the shift to the change. But you have no way of knowing whether the change caused the movement, whether it was a coincidence, whether an algorithm update was the real driver, or whether a competitor simply lost a link. Without documentation, every win is a guess and every loss is a mystery.

This is why the SEO journal matters. Not as a productivity ritual, but as an epistemological tool. It forces you to write down what you believe before the result arrives, which is the only way to test whether your beliefs are actually correct.

If you want a broader framework for how this kind of disciplined thinking fits into a full search strategy, the Complete SEO Strategy hub covers the end-to-end picture, from technical foundations to content architecture to measurement.

What an SEO Journal Actually Contains

The format matters less than the consistency. I have seen teams use Notion, Google Sheets, Confluence, and plain text files. All of them work. What distinguishes a useful journal from a useless one is not the tool, it is whether the entries contain enough information to be meaningful six months later.

A well-maintained SEO journal has five components.

1. The Change Log

Every material change to your site that could affect search performance should be recorded with a date. This includes content updates, title tag rewrites, internal link additions, page merges, URL changes, schema additions, Core Web Vitals fixes, and robots.txt amendments. It also includes things you did not do intentionally: a developer pushing a noindex tag to production, a CMS update that altered your URL structure, a CDN configuration that broke your sitemap.

The change log is your alibi. When rankings drop, you look at what changed in the preceding four to six weeks. When rankings improve, you do the same. Without the log, you are guessing. With it, you have a starting point for a real investigation.

2. The Hypothesis

Before you make a change, write down what you expect to happen and why. This does not need to be elaborate. A single sentence is enough: “Rewriting this title tag to include the modifier ‘for small businesses’ should improve click-through rate on this query cluster because the current title is too generic for the audience.”

The discipline of writing a hypothesis before acting is the single most underused practice in SEO. It forces you to think about why a tactic should work, not just whether it is commonly recommended. It also gives you something to evaluate when the results come in, rather than retrofitting a narrative to whatever happened.

3. The Observation

Four to eight weeks after a change, record what you observed. Was the hypothesis correct? Partially correct? Wrong? Did something unexpected happen? Did a different page benefit more than the one you changed? Did rankings move but traffic stay flat, suggesting a click-through rate problem rather than a positioning problem?

The observation entry is where most of the learning lives. It is also where most teams stop keeping the journal, because writing down “this did not work” feels like admitting failure. In reality, a documented failure is worth more than an undocumented success, because it stops you from repeating the same mistake.

4. The External Context

SEO does not happen in isolation. Google updates its algorithm continuously, with major confirmed updates occurring several times per year. Competitors gain and lose links. Seasonal search behaviour shifts. Economic conditions change what people are searching for. Your journal should include a note of any confirmed algorithm updates during the period you are reviewing, as well as any significant competitive movements you are aware of.

This context layer is what separates a useful journal from a simple change log. It lets you distinguish between “our rankings dropped because we made a mistake” and “our rankings dropped because the entire industry moved down during a broad core update.” Those two scenarios require completely different responses.

5. The Forward Notes

At the end of each monthly review, write down two or three things you want to test or investigate in the next period. These are not tasks. They are questions. “Why is this page ranking for queries that have nothing to do with its topic?” or “Why did this cluster of pages lose 30% of their impressions in the last six weeks while rankings held steady?” The forward notes keep the journal from becoming a passive record and turn it into an active research programme.

The Cadence That Makes It Work

I have watched teams try to maintain daily SEO journals and abandon them within a fortnight. The overhead is too high, the signal-to-noise ratio in daily data is terrible, and the exercise becomes a chore rather than a tool. Daily is the wrong cadence for SEO, full stop.

Weekly entries work for the change log, because you want to capture what happened while it is still fresh. But weekly reviews of ranking data introduce too much noise. A page can swing three positions in a day for reasons that have nothing to do with your work, and if you are reviewing weekly, you will start chasing ghosts.

Monthly is the right cadence for analysis. Pull your ranking data, your traffic data, and your Search Console impressions and clicks on a monthly basis. Compare month-on-month and year-on-year. Identify the pages that moved materially in either direction. Cross-reference against your change log and your external context notes. Then write your observations and your forward questions.

Quarterly is the right cadence for strategy. Every three months, read back through your monthly entries and look for patterns. Are there content categories that consistently outperform? Are there technical issues that keep resurfacing? Are there hypotheses that have been consistently wrong, suggesting your mental model of how this site or this industry responds to SEO is off?

The quarterly read-back is where the real value accumulates. It is also where you will notice things that are invisible at the monthly level: slow content decay on pages that were strong eighteen months ago, a gradual shift in which queries are driving traffic as search behaviour evolves, a pattern of algorithm updates that consistently affect your site in a particular way.

What the Journal Reveals That Dashboards Cannot

Dashboards show you what is happening. Journals show you why, or at least give you a structured way to investigate why. That distinction matters more than most teams realise.

I spent a significant portion of my agency career sitting in client review meetings where someone would point at a traffic chart and say “this went up” or “this went down” as if the observation itself was analysis. It was not. The chart was a symptom. The journal is the diagnostic tool.

There are four specific things a journal surfaces that dashboards consistently miss.

Content decay patterns. Dashboards show you current performance. Journals show you trajectory. A page that is performing adequately today but has been declining for eight months is a different problem from a page that dropped sharply in the last six weeks. The journal lets you see the slow-burn decay before it becomes a crisis, because you have been recording impressions and clicks against that page in every monthly entry.

The lag between action and result. SEO changes rarely produce immediate results. A content update might take six to ten weeks to be fully re-crawled, re-indexed, and re-evaluated. A link acquisition might take weeks to influence rankings. If you are only looking at dashboards, you will often attribute a result to the wrong cause, because you are looking at what happened recently rather than what happened two months ago. The journal’s change log, cross-referenced against your observation entries, makes the lag visible.

The compounding effect of small changes. Individual SEO changes are often too small to produce statistically meaningful results on their own. But ten small changes made over six months can produce a significant cumulative effect. Without a journal, you cannot see this compounding. You made ten changes, traffic improved, and you have no idea which combination of changes drove the improvement. With a journal, you can at least narrow it down and form more informed hypotheses about which interventions are doing the most work.

The relationship between technical health and content performance. Some of the most instructive patterns I have seen in client SEO data involve technical issues affecting content performance in non-obvious ways. A crawl budget problem that was suppressing a high-value content cluster. A canonical tag issue that was splitting authority between two versions of the same page. These patterns only become visible when you are tracking both technical changes and content performance in the same document over time. The journal creates that cross-domain visibility.

Building the Habit Without Building a Bureaucracy

The biggest risk with any documentation system is that it becomes the work rather than supporting the work. I have seen teams spend more time maintaining their SEO tracking systems than actually doing SEO. That is the wrong trade-off.

The journal should take no more than thirty minutes per week for the change log entries, and no more than two hours per month for the analysis review. If it is taking longer than that, the format is too complex.

Start with the minimum viable version. A shared Google Doc with three sections: changes made this week, what I expect to happen and why, and what I observed from last month’s changes. That is it. No elaborate templates, no colour-coded spreadsheets, no custom dashboards. Just a consistent record of decisions and observations, maintained with discipline over time.

The sophistication can come later, once the habit is established. Add the external context layer when you have been maintaining the journal for three months. Add the quarterly review process at the six-month mark. Build complexity only when the simpler version is already working, because complexity added too early is the most common reason documentation systems fail.

For teams working across multiple sites or clients, the journal structure scales reasonably well. The key adjustment is adding a site or client identifier to every entry, and being disciplined about not mixing observations across different properties. A ranking shift on one site tells you nothing about another site, even if they are in the same industry, and conflating them is a fast route to bad conclusions.

The Honest Limitations of Keeping a Journal

I want to be clear about what a journal cannot do, because overselling any tool is a habit I have tried to break after spending too many years watching agencies oversell their methodologies to clients.

A journal cannot give you certainty about causation in SEO. Google’s algorithm has hundreds of ranking factors, many of which interact in ways that are not publicly documented. You will make changes, observe results, form hypotheses, and some of those hypotheses will be wrong. The journal makes you less wrong over time by giving you a structured way to test and revise your beliefs. It does not make you right.

A journal cannot replace competitive intelligence. Your rankings are partly a function of what your competitors are doing, and your journal only captures your own activity. If a competitor acquires a significant number of links, refreshes their content architecture, or launches a new content programme, your rankings will be affected regardless of what you did or did not do. The external context section of the journal helps, but it is a partial solution at best.

A journal cannot compensate for a fundamentally weak SEO strategy. If you are targeting the wrong keywords, producing thin content, or operating on a technically compromised site, documenting your activity more carefully will not fix those problems. The journal is a learning tool, not a substitute for sound strategy. If you are still working through the strategic foundations, the Complete SEO Strategy hub is a good place to work through the fundamentals before optimising your documentation process.

What the journal does, done consistently over twelve to eighteen months, is give you something genuinely rare in SEO: a defensible account of your own decision-making. That is worth having, both for your own learning and for the conversations you will inevitably have with stakeholders who want to know why performance moved in a particular direction.

A Note on Using AI in Your SEO Journal Process

AI tools are increasingly being used to assist with SEO content and analysis, and it is worth addressing how they fit into a journal-based workflow. The short answer is that they are useful for some parts of the process and actively counterproductive for others.

AI is useful for drafting content at scale, generating title tag variations, and identifying patterns in large datasets. Moz has covered the practical applications of generative AI for SEO content in reasonable depth, and the honest assessment is that AI accelerates execution without replacing strategic judgment.

Where AI is counterproductive in a journal context is in the analysis and hypothesis stages. AI tools will give you plausible-sounding explanations for ranking changes, but those explanations are generated from patterns in their training data, not from knowledge of your specific site, your specific competitive environment, or the specific changes you made. Using AI to write your observations and hypotheses is a fast route to confident-sounding nonsense.

The journal is a thinking tool. The thinking has to be yours. AI can help you organise information, draft content, and identify anomalies in data. The interpretation, the hypothesis formation, and the strategic conclusions need to come from a human who understands the context. That is not a romantic view of human cognition. It is a practical observation about where AI currently adds value and where it introduces noise.

What Consistent Documentation Does to Your SEO Career

There is a professional dimension to this that does not get discussed enough. SEO practitioners who maintain detailed journals of their work over years develop a depth of pattern recognition that is genuinely difficult to replicate through other means. They have seen what works in their specific industry, on their specific type of site, for their specific audience, across multiple algorithm cycles. That accumulated knowledge is not available in any course, any blog post, or any tool.

Moz has made the point that SEO career development increasingly requires practitioners to demonstrate strategic thinking rather than just tactical execution. A journal is one of the most concrete ways to develop and demonstrate that strategic thinking, because it forces you to engage with the question of why, not just what.

I have interviewed a lot of SEO practitioners over the years, and the ones who stood out were almost always the ones who could talk about specific experiments they had run, what they expected, what actually happened, and what they concluded. That level of specificity only comes from documentation. It cannot be faked in an interview, and it cannot be acquired by reading other people’s case studies.

Search Engine Journal has noted that the most durable SEO careers are built on genuine expertise rather than familiarity with current best practices, because best practices change while expertise compounds. A journal is one of the primary mechanisms through which expertise compounds. You are not just doing the work. You are building a record of the work that becomes more valuable the longer you maintain it.

That is not a small thing. Most of what passes for SEO knowledge is either generic advice that applies to everyone and therefore helps no one specifically, or platform-specific tactics that will be obsolete within eighteen months. A well-maintained journal of your own work, in your own industry, on your own sites, is neither of those things. It is specific, it is current, and it belongs entirely to you.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions