A practical guide to SEO: strategy, hiring, and growth levers

Executive overview

Most companies under-invest in SEO relative to paid ads, despite comparable traffic potential. The core problem is that SEO is treated as a technical audit exercise rather than a growth strategy rooted in product and editorial quality.

The path to winning at SEO is knowing your addressable market, matching page types to search intent, building topical authority, and ensuring internal link coverage — then getting things launched.

When SEO makes sense

• Minimum signal: ~1,000 non-SEO visits/day (direct, paid, social) — proves domain credibility to Google
• Minimum signal: ~1,000 referring domains (check via Ahrefs or SEMrush)
• Size the market by looking at product competitors and audience competitors in SimilarWeb, then multiply traffic by estimated conversion rate
• Compare expected SEO return against other channels (paid, social) before committing resources
• Starting from zero with no authority: too early — wait until organic traction appears

The three types of SEO

• Programmatic SEO: pages auto-generated from a database — product pages, address pages, category grids (e.g. Zillow, eBay)
• Editorial SEO: human-written articles, guides, listicles — now drives more traffic than programmatic for most companies
• Technical SEO: internal link architecture, redirects, page speed, tags — foundational but not a growth lever on its own
• Editorial SEO is available to every company; programmatic SEO is an additional option where UGC or large inventories exist
• Programmatic has low per-page cost at scale; editorial has a cost per page but higher flexibility

Keywords vs. topics

• Google's algorithm targets topics, not individual keywords — one page typically ranks for 200–2,000 keyword variations
• Treating keywords as topics understates search volume by ~10x and leads to wasted content spend
• Cluster keywords by checking how much overlap exists in Google results — identical results = same topic, same page
• For each topic, identify subtopic themes by auditing competing pages in Ahrefs/SEMrush and checking for gaps in your content
• Missing subtopics = lower content score = worse engagement = lower rankings

Determining the right page type

• Take target keywords, search them in Google, and look for patterns in what page types rank
• If articles rank, you need articles; if category grids rank, you need category pages — a mismatch will not rank
• Look at what components appear consistently on high-ranking pages (maps, phone numbers, templates) — these signal user intent
• Build pages that genuinely serve that intent; better engagement directly lifts rankings

Content and engagement signals

• Google tracks click-through rate from search results and whether users return to Google after visiting (pogo-sticking is a strong negative signal)
• Time-on-page and session completion are tracked via Chrome and Android — a page that ends the search session ranks better
• Comprehensive topic coverage drives higher content scores and longer dwell time
• Engagement improvement is an immediate ranking signal, not just a long-term benefit

How Google measures authority

• Links: referring domain count is the baseline signal
• Non-SEO traffic: direct, paid, social — signals to Google you are a real brand, not an SEO farm
• Branded search: frequency of people typing your brand name into Google
• Shares and social activity: what content gets shared and the anchor context
• Authority is topical, not just domain-wide — being disproportionately known for a narrow theme lets you outrank larger domains on that theme
• Adjacent topics benefit from topical authority wins (e.g. a jello shots post boosted rankings for all alcohol-adjacent recipes across the domain)
• Start from what you already have authority for; branch out to adjacent topics from there

Internal links: the underused lever

• ~90–95% of sites have this problem: a power-law distribution where 5% of pages get 95% of internal links
• Google crawls via links — pages with few internal links are crawled less often and carry less authority
• Target at least 5–10 internal links per page; pages with zero or one link are effectively invisible
• Standard algorithms (popular, recent, related, faceted) all skew links toward a subset of pages and leave others starved
• Fix: build a custom internal link layer that distributes links broadly, in the page body (not just footer or nav)
• Body links count significantly more than nav/footer links
• Expected lift from fixing internal links: 25–100%+ traffic improvement

Timelines and leading indicators

• Typical timeline from zero: 3 months to build and launch, then 3–6 months to see ranking movement
• Large, authoritative domains may see results in 1–4 weeks after optimization
• Leading indicator: pages moving from position 15 → 12 → 8 → 5 for target keywords — signals the strategy is working
• Scale investment only after seeing these indicators from an initial batch of 10–30 articles
• Common mistake: concluding failure too early, or testing before enough content is launched

Tools

• Google Search Console (free): real traffic data, keyword rankings, SERP tracking
• Clearscope: content scoring and subtopic gap analysis — shows what your content is missing
• Ahrefs or SEMrush (~$100–200/month): keyword research, competitor URL and keyword audits, referring domain data
• Screaming Frog (~$150/year): site crawl, internal link count audit, redirect and error detection
• SimilarWeb: estimate competitor traffic by channel to size addressable market

Hiring for SEO

• Programmatic SEO: specialized skill — look for people who achieved high SEO traffic percentage at lesser-known companies (harder route, less competition for them)
• Editorial SEO: more accessible — a motivated generalist can learn the workflow; domain-expert writers handle content, SEO manager handles topic selection and outlines
• Minimum editorial team: one SEO manager + one writer
• Bias toward in-house for both; agencies can be useful early or when internal resources are blocked
• Typical volume to start: 10–25 articles/month, scaling based on ROI signals

Getting things done

• The most common failure mode: strategy is defined but almost nothing gets launched
• Root causes: SEO sits outside the product org, low executive buy-in, perceived mystery around outcomes, and high scope of work
• Fix: embed SEO within product prioritization, secure executive sponsorship, and frame decisions as impact-vs-scope trade-offs
• Getting things launched matters more than strategy refinement

AI and SEO

• AI-generated content without underlying structured data produces fluent but factually unreliable sentences — not recommended for health, finance, or high-stakes topics
• Useful application: AI for workflow steps — topic clustering, subtopic extraction, content scoring, gap analysis
• Useful application: structured-data-driven generation (e.g. sports scores, clinical trial fields) where facts are known and machine-readable
• Human writers with domain expertise should write the final content; AI should inform the structure and priorities