Content Strategy & Editorial Planning

A content strategy is not a backlog of “ideas.” It is an explicit plan that says which clusters you’ll dominate, which briefs writers will execute against, and how you’ll measure when a piece is actually finished. Without those three artifacts, your content team is just rolling dice.

TL;DR

• Pillar-cluster (hub-and-spoke) is the default architecture. A pillar page covers the topic at length and links to 10–40 cluster pages that cover specific subtopics; clusters link back to the pillar. This is how Google reads your site as one trusted source on the topic.
• The brief is the deliverable, not the article. A complete brief includes target query, intent, SERP feature target, outline, entity list, internal links to add, and a scorecard. Ranking content comes from briefs that take 90 minutes to write, not from writers guessing.
• Gap analysis beats ideation. Every quarter, run a competitor content gap, score what you have against what ranks, and prune or rewrite the bottom decile. The compounding effect of pruning is larger than publishing more.

The mental model

A content strategy is like a railway timetable: useless without two things — the map of stations (your topical architecture) and the schedule of trains (the editorial calendar). Most teams have one or the other, never both.

The map is your pillar-cluster diagram: a tree of pillars (broad topics, evergreen) with clusters underneath (specific queries, more time-sensitive). The schedule is the editorial calendar: which cluster article ships on which date, which writer is assigned, and which pillar it strengthens. Without the map, the schedule turns into random posts. Without the schedule, the map is a wishlist.

The brief is what links the two. It tells a writer not “write 1,500 words on X” but “produce the article that wins position 1 for this query, satisfies this intent, captures this SERP feature, links back to these three pages on our site, and scores ≥85 on our scorecard before you submit.”

Deep dive: the 2026 reality

The pillar-cluster model is not new — HubSpot popularized it in 2017 — but the scoring layer is what separates 2026 production from 2018 amateur work.

Modern content briefs include four numerical targets:

1. Word count band, derived not from “longer is better” but from the median length of the top three ranking results. As of 2026, a query with three top results averaging 1,400 words gets a brief targeting 1,300–1,600. Going to 4,000 words on a 1,400-word SERP signals padding, which the Helpful Content system has explicitly flagged since the September 2023 update.

2. Entity coverage score — how many of the entities present across the top 10 results does your draft mention? Tools like Surfer SEO, MarketMuse, Frase, and Clearscope all compute this; the practical floor in competitive niches is 80% coverage.

3. Readability, measured against the audience. Flesch Reading Ease ≥60 for consumer topics, 40–60 for B2B, 30–50 for technical. Hemingway’s “Grade 9 or below” is the right target for SaaS comparison pages; medical content can sit at Grade 11 if the SERP supports it.

4. Original content ratio — the percentage of the article that is not paraphrased from competitors. Tools like Originality.ai and GPTZero flag derivative writing; the March 2024 core update specifically targeted sites whose content was 90%+ rewordings of better sources.

For AI surfaces, the math is different. AI Overviews, AI Mode, Perplexity, ChatGPT Search, and Claude with web all extract claims from articles and cross-reference them. The articles that get cited are the ones with load-bearing original facts — original data, expert quotes, primary research — not the ones with the highest entity coverage. Your editorial calendar should allocate ~20% of slots to original-research pieces, even though they cost 5–10× a synthesis post; they’re what gets cited.

The competitor gap analysis has also evolved. In 2018, you’d export Ahrefs Content Gap and write everything competitors ranked for. In 2026, that produces a Helpful Content penalty. The 2026 workflow: pull the gap, score each missing query against your brand fit, kill the ones that don’t fit, and only then add the survivors to the calendar.

Visualizing it

flowchart TD
  A["Pillar: Email Marketing"] --> B["Cluster: Subject lines"]
  A --> C["Cluster: Deliverability"]
  A --> D["Cluster: Segmentation"]
  A --> E["Cluster: Automation"]
  B --> B1["Best subject line length"]
  B --> B2["Subject line A/B testing"]
  C --> C1["SPF, DKIM, DMARC setup"]
  C --> C2["Avoiding spam folder"]
  D --> D1["RFM segmentation"]
  D --> D2["Behavioral triggers"]
  E --> E1["Welcome series design"]
  E --> E2["Abandoned cart flows"]
  B1 -.back to pillar.-> A
  C1 -.back to pillar.-> A
  D1 -.back to pillar.-> A

Solid arrows are pillar-to-cluster links; dotted arrows are the cluster-back-to-pillar links that complete the loop.

Bad vs. expert

The bad approach

The classic failure: a content brief that reads like a mood board.

# Brief: Email subject lines
Topic: subject lines for email marketing
Length: ~1500 words
Tone: friendly, expert
Keywords: email subject lines, best email subject lines, subject line tips
Outline:
- Intro
- Why subject lines matter
- Tips
- Examples
- Conclusion
Due: Friday

This produces a generic article that ranks nowhere. There is no target query, no intent classification, no SERP analysis, no entity list, no internal-link plan, and no scorecard. The writer makes 50 small decisions on their own, and 49 of them dilute the page.

The expert approach

A production-grade brief is structured, machine-readable, and reusable.

brief_id: "EM-014-subject-lines-length"
target_query: "best email subject line length"
search_volume: 2400
keyword_difficulty: 28
search_intent: "informational"
serp_feature_target: "featured_snippet_paragraph"
serp_analysis:
  top_3_avg_word_count: 1380
  top_3_avg_h2_count: 8
  top_3_dates_published: ["2024-11", "2025-03", "2025-09"]
  paa_questions:
    - "What is the ideal subject line length?"
    - "Does subject line length affect open rate?"
    - "How long should mobile subject lines be?"
target_word_count: [1300, 1500]
target_flesch_score: 65
required_entities:
  - "open rate"
  - "preview text"
  - "Litmus"
  - "Gmail mobile clipping"
  - "MailChimp benchmark report"
required_data:
  - "Original analysis of 10,000+ emails or cite source with date"
internal_links_to_include:
  - "/email/deliverability/"
  - "/email/preview-text/"
  - "/email/ab-testing/"
schema:
  - "Article"
  - "FAQPage"
scorecard_threshold: 85
author_qualifications:
  - "Has run email campaigns at $1M+ ARR"
due_date: "2026-05-22"

The brief encodes intent (“informational”), the SERP feature you’re trying to win (“featured_snippet_paragraph”), the data the article must contain, and which internal pages it must link to. A scorecard with a numeric threshold turns “is this done?” from an opinion into a measurement.

Brief element	Bad	Expert
Target query	”subject lines” (head term)	One specific query with intent
Length target	”around 1500”	Range derived from SERP median
Internal links	Not mentioned	Listed by URL with anchor guidance
Data	Not required	Original or cited primary source
Done definition	Editor’s gut feel	Numeric scorecard threshold

Do this today

1. Open Ahrefs or Semrush. In Site Explorer, enter your top competitor and run Content Gap against your domain. Filter to keywords with KD ≤ 40 and monthly volume ≥ 200. Export the result.
2. In a spreadsheet, tag every gap query with one of: Pillar, Cluster, Outer, Skip. “Skip” is for queries off your brand strategy. Aim for 20–30% Skip.
3. Map every surviving query to a cluster. If a query doesn’t fit an existing cluster, you may need a new one — but resist the urge to create more than three new clusters per quarter.
4. In Notion, Airtable, or Google Sheets, build an editorial calendar with columns: Brief ID, Pillar, Cluster, Target Query, Writer, Editor, Status, Due Date, Publish Date. Use the same Brief ID format as your filename (“EM-014-subject-lines-length.md”) so commits link to briefs.
5. For every brief, run the target query in an incognito Chrome window and capture the top 3 word counts (use Word Counter Plus extension), the top 5 People Also Ask questions, and the rich features present (FAQ, video, snippet, image pack). Paste these into the brief’s serp_analysis block.
6. Add a content scorecard to your CMS or PR template. Fields: word count in band (yes/no), flesch score (number), required entities mentioned (count/total), original data present (yes/no), internal links present (count/required), schema valid (yes/no via Schema Markup Validator). Threshold to publish: 5 of 6.
7. Schedule a monthly content audit on the calendar. Pull GSC’s Performance report for the past 90 days, sort URLs by clicks, and flag the bottom 20% that have been live ≥6 months — these are decay candidates for module 17’s prune/refresh workflow.
8. Quarterly, run a competitor content gap again. The list will shrink as you publish — when it stops shrinking, switch from publishing-mode to refresh-mode.
9. For every pillar, dedicate one calendar slot per quarter to an original-data piece (survey, log analysis, dataset). Allocate 5–10× the production budget; this is what gets cited in AI Overviews and Perplexity.
10. In Looker Studio (or your BI of choice), build a cluster-level dashboard that pivots GSC data by URL prefix. Track clicks, impressions, average position, and pages-per-cluster month-over-month. The cluster, not the page, is your unit of optimization.