SEO Decision Log (Ongoing)

Live project: uporbit.app

One-Line Summary: Built UpOrbit as a personal productivity and organization system after realizing most tools optimize for features rather than actual use, using iterative design, behavioral insights, and lessons learned from previous SEO and systems-based projects to create something simple enough to use daily while exploring scalable product architecture across web, extension, and mobile environments.

Context: The project began from a practical frustration rather than a market analysis. Most productivity tools are powerful but rarely used consistently. Features accumulate faster than habits form, and over time, the tool itself becomes another source of friction. The initial question became simple: would this be something I actually use every day? Every design decision ultimately traced back to that constraint.

Instead of building for theoretical users, the system was designed around personal workflow first. Tasks, reminders, progress tracking, and organization were evaluated based on whether they reduced mental overhead or added to it. Features that required explanation or repeated setup were removed or simplified. The goal was not maximum capability, but sustained usability.

At the same time, the project expanded into a broader theme that now carries into the associated blog. The intent is not to present medical or psychological authority, nor to prescribe solutions. The framing is practical and experiential. The underlying idea is simple: people are trying to improve their lives, and small structural improvements in organization, reminders, and decision clarity can have a meaningful impact over time.

Development followed a familiar pattern from earlier projects. Early versions were discarded repeatedly after revealing unnecessary complexity or unclear interaction patterns. The home screen, task flow, and notification systems were redesigned multiple times to reduce friction and make priorities visible without overwhelming the user.

A major component of the project involved expanding technical capability beyond previous work. Building UpOrbit required learning new environments and constraints, including Chrome extension architecture, mobile-oriented interaction patterns, and synchronization considerations between local and account-based storage.

Lessons from the water filtration project were carried directly into this work. Project management structure, template thinking, and organizational discipline became more important as complexity increased. Components were designed to be reusable, interfaces standardized, and features evaluated at the system level instead of as isolated additions.

Options considered: Several directions were explored, including adapting existing productivity tools through integrations or focusing purely on content and education rather than building software directly. Those approaches were abandoned because they did not solve the original problem.

Rationale: A tool intended to improve productivity must justify its own existence through repeated use. Designing around personal adoption creates a stricter constraint than designing around feature lists. By prioritizing simplicity, clarity, and low friction, the system has a better chance of becoming part of routine behavior rather than another abandoned application.

Outcome: The current version establishes a functional foundation across web and extension environments, with improved interconnectivity between tasks, reminders, and daily priorities. The system is usable in its current state, which serves as the primary validation metric. Ongoing work focuses on refinement rather than expansion.

Biggest Challenges and Risks: The primary risk is overbuilding. Productivity tools naturally trend toward complexity, and maintaining restraint requires continuous evaluation of whether new features solve real problems or simply add novelty. The technical scope also expands quickly when supporting multiple platforms, increasing the importance of disciplined project management and clear architectural decisions early.

Live project: checkmytap.com

One-Line Summary: Built a water filtration and softener comparison platform after identifying a clear market gap, investing significant time in prompt engineering, competitive analysis, and scalable template design to create a system capable of delivering genuinely useful information at scale while testing modern SEO workflows and development tooling.

Context: The project began with a broader evaluation of potential niche sites where structured information, high buyer uncertainty, and weak existing competition created an opportunity for long-term organic growth. Water filtration and softeners stood out due to a combination of factors. Existing competitors were largely outdated in structure, difficult to navigate, and often optimized around aggressive monetization rather than clarity or usefulness.

This created a clear gap between what users needed and what existing sites provided. The opportunity was not simply traffic acquisition, but improving how information is organized and presented so that users could make confident decisions.

A substantial portion of the work occurred before the site was publicly visible. Significant time was spent on prompt engineering, refining how structured content could be generated consistently without sacrificing accuracy or intent alignment. Templates were iterated repeatedly to remove redundancy, reduce ambiguity, and ensure each page served a distinct purpose within the broader site architecture.

The development process emphasized systems over individual pages. The goal was to create templates that could scale without degrading quality. Many early approaches were discarded after revealing structural weaknesses or unnecessary complexity. Each iteration improved clarity around what should be automated, what required human oversight, and how to balance efficiency with responsibility.

Prompt engineering became a core skill within the process. Rather than treating AI tools as content generators, they were used as structured assistants guided by detailed constraints, competitive context, and explicit SEO objectives.

Options considered: Many. There are roughly 30 versions of the site.

Rationale: A scalable system built early makes it possible to grow without repeatedly revisiting foundational decisions. Instead of relying on copy variation to differentiate pages, the approach emphasizes structural clarity: consistent page types, stable URL patterns, tight internal linking logic, and modules that map to real user questions. Templates are standardized so every new page inherits the same crawlability, content hierarchy, schema readiness, and UX signals.

Outcome: The project launched with standardized page structures for comparisons and solution education, with consistent internal linking and content modules designed to reduce ambiguity for both users and crawlers. Currently in the wait-and-see phase, monitoring how Google treats the site over time.

Biggest Challenges and Risks: Creating content that is unique on a per-page basis, adding real value rather than scaling low-quality output. Balancing affiliate links carefully without losing credibility or trust. Learning new ways of project management as files grow larger and the project scales.

Related: Full case study

Context: Google Search Console is commonly used as a troubleshooting tool rather than a visibility analysis tool. Most reporting relies on averaged metrics across fixed periods, which compresses meaningful movement into summary numbers. This makes it difficult to explain momentum, volatility, or the impact of specific actions such as launches, structural changes, or content publication. As AI-generated summaries and answer-first search reduce direct clicks while impressions increase, visibility trends require clearer interpretation.

Problem: Monthly or quarterly summaries hide turning points and reduce visibility into how performance actually evolves. Strategists and clients often see flat or inconsistent metrics without understanding whether underlying visibility is improving, declining, or stabilizing.

Rationale: Instead of analyzing only aggregated data, export day-by-day GSC performance data across defined time windows and use AI analysis to identify patterns, sustained trends, and meaningful changes in visibility. The output format is controlled based on audience type, allowing the same dataset to support client-facing summaries, internal strategy discussions, or technical analysis.

Reasoning: Day-level data preserves sequence and momentum, allowing trend identification that is invisible in averaged reports. AI tools are effective at identifying patterns across longer datasets and calculating percentage change consistently, reducing manual interpretation time while improving clarity in communication.

Risks: AI-generated analysis may overfit patterns or imply causation without context if prompts are not controlled carefully. GSC data isn’t always accurate, and has gone through some changes (especially the mid-July update).

Status: Active. Ongoing evaluation as reporting cycles accumulate.

Related: Why impressions increase while clicks stay flat

One-Line Summary: Led SEO strategy across five 3.0 site launches for a large U.S.-based heavy equipment manufacturer by working closely with developers, product managers, and external teams to standardize technical foundations, content structure, and post-launch validation, reducing launch risk and supporting long-term scalability.

Context: The organization was rolling out multiple 3.0 site launches for a large heavy equipment manufacturer operating across the western United States and beyond. The sites included complex equipment inventory, category and model-level pages, multi-state and city-level dealer location structures, parts and service sections, and a large volume of legacy URLs. Content flexibility was limited by manufacturer standards, shared specifications, and compliance requirements.

Options considered: Treat each launch independently or apply a shared SEO system across all site launches.

Rationale: A shared SEO system made it possible to scale launches without repeating the same technical and structural issues. Working closely with developers and product managers ensured SEO requirements were built into templates, equipment listings, content widgets, and location structures from the start. Pre-launch audits focused on identifying systemic risks across equipment inventory, dealer locations, parts, and services.

Outcome: Across five launches, audits helped catch and fix common issues like 404s, broken links, duplicate content, pagination problems, and inconsistent page structure before they turned into bigger problems. Equipment category and model pages shipped with consistent templates and content modules, and location pages followed a clear state and city structure. After launch, crawl behavior and indexation stayed stable, with no major visibility drops.

Biggest Challenges: Balancing client requests with SEO best practices, especially when requested changes conflicted with crawlability, structure, or long-term scalability. Coordinating with developers and PMs on complex structural changes across inventory, location, and template logic. Migrating legacy content into the new 3.0 framework without losing relevance, introducing duplication, or breaking existing search equity.

What is reinforced: Enterprise SEO for heavy equipment manufacturers is a systems and coordination problem, not a page-by-page exercise. Close collaboration with developers and product managers reduces risk more effectively than post-launch fixes.

Related: Building repeatable SEO systems for enterprise site launches

One-Line Summary: Decided to consolidate related topics into high-quality pillar pages and supporting sections rather than publishing isolated posts, so search systems and users can understand the full idea in one place.

Context: As the site grew, several ideas naturally overlapped. Publishing each idea as a separate post risked fragmenting context, duplicating intent, and making it harder for search systems to understand how concepts connect. Modern search systems increasingly favor coherent topical coverage over scattered single-use pages.

Options considered: Publish each idea as a standalone post, rely on tags and categories to connect related content, manually interlink posts after publishing, or consolidate related ideas into pillar pages with supporting sections.

Rationale: A pillar-first approach reflects how people actually learn and research. Users want to understand a concept fully, not hop between disconnected posts. For modern search systems, consolidated pages provide clearer signals around topic ownership, relationships, and intent. Supporting sections and internal links preserve depth without relevance loss.

Outcome: Pillar pages are driving stronger engagement, longer time on page, and broader query coverage than isolated posts. Related queries are being captured on fewer, stronger URLs, reducing internal competition and improving overall clarity for both users and search systems. Still using posts for long-tail searches to give the pillar pages more juice.

What is reinforced: Modern search favors clarity over volume. Fewer, stronger pages that fully explain an idea are easier for both humans and AI systems to interpret. Content architecture is no longer just organization, it is a ranking and discovery signal.

One-Line Summary: Added long-tail FAQ sections to key pages to capture real search questions, align with how modern search systems interpret intent, and improve visibility across AI-driven and traditional search results.

Context: Modern search behavior is increasingly question-driven, fragmented, and influenced by AI systems that extract, summarize, and recombine information. Many key pages were strong conceptually but did not directly answer the specific, long-tail questions users actually ask during research and decision making. This limited visibility in AI Snippets, AI Overviews, and long-tail SERP features, even when the core content was relevant.

Options considered: Leave pages as-is and rely on primary content to rank, create separate blog posts for each long-tail question, add FAQ sections focused on generic high-volume questions, or add targeted long-tail FAQ sections directly to key pages.

Rationale: Adding long-tail FAQs directly to key pages allows the content to match how people actually search without forcing them into separate posts. These sections help search systems understand intent, relationships, and coverage more clearly, especially for AI-driven discovery. Placing FAQs on existing pages strengthens topical depth and supports both traditional rankings and AI extraction without fragmenting the site.

Outcome: Early signals show increased impressions across long-tail queries and improved visibility in AI Snippets and AI Overview results. Pages are surfacing for a wider range of specific questions, suggesting stronger alignment with modern search behavior.

What is reinforced: Modern search rewards pages that clearly answer real questions, not just broad topics. Long-tail FAQs help bridge the gap between intent, content, and AI interpretation. Structuring answers where users already are improves discoverability without adding unnecessary content sprawl.

Related: The real goal of technical SEO: clarity for crawlers and humans

One-Line Summary: Built a comparison pillar and supporting competitor pages to capture comparison searches, answer real buyer questions, and guide users naturally from research into available INEOS Grenadier inventory. The pages were also structured to support local SEO by reinforcing dealership relevance and inventory availability.

Context: Shoppers in the outdoor 4×4 market usually compare vehicles before committing to a brand. This INEOS Grenadier dealership lacked clear comparison content that matched real search behavior, had limited visibility relative to established U.S. manufacturers, and did not provide a smooth path from research to available inventory. Local relevance was also underutilized in comparison-driven searches.

Options considered: Do nothing and rely on brand/inventory pages, create a single comparison page covering all competitors, create individual comparison pages without a central pillar, or build a comparison pillar supported by individual competitor pages linked to inventory.

Rationale: This approach shows up where people are already comparing options instead of waiting for them to search the brand directly. The individual comparison pages meet very specific searches, and the main page ties everything together. By explaining differences first and introducing inventory later, the pages help people decide without feeling pushed.

Outcome: In three months, six total pages have already generated 226 clicks and 39.8K impressions, with an average position of 5.9. AI Snippets and AI Overview mentions are increasing sitewide. New inventory pages are also seeing stronger search performance, with impressions up 34 percent, clicks up 6 percent, and average position improving by 4.2 places.

What is reinforced: Comparison content works best when it mirrors how people actually research, not how brands want to be discovered. When combined with local SEO signals and inventory context, comparison pages can influence visibility beyond the comparison itself. Strong internal linking and local relevance help inventory surface earlier in the decision process.