The SEO gap analysis playbook: how to find the exact posts your competitor is winning with (as of July 2026)

Step 1: Pick 5 direct competitors, not 20

5 is the sweet spot: real patterns, sub-10-min indexing. 20 gives you mush and quadruples the work.

The instinct is to list every SaaS in your category. Resist it. Five direct competitors is the sweet spot, enough to see real patterns, few enough that indexing finishes in under 10 minutes.

Pick criteria:

1. 1Same primary use case.
2. 2Same buyer persona.
3. 3They've published at least 20 blog posts.
4. 4Their homepage ranks in the top 20 for your head-term.

Step 2: Index their content

Fetch each sitemap, extract blog URLs, cluster into 8-15 topic buckets. Eyeball-cluster is 80% as good as embeddings in 15 min.

For each competitor, fetch the sitemap. It lives at /sitemap.xml on ~95% of SaaS sites. If it doesn't, try /sitemap_index.xml or check /robots.txt (the sitemap location is usually declared there). WordPress, Ghost, Next.js, and Astro all expose sitemaps by default.

Extract every URL matching /blog/*, /posts/*, /articles/*, or /guides/*. That's your candidate set. For most SaaS blogs you'll pull 30-150 URLs per competitor.

Note next to each bucket: how many URLs contribute, and which competitor dominates it. Dominance matters for step 5.

Step 3: Index your own content the same way

Use the competitor bucket list as your vocabulary. Do not invent new buckets. Unfittable posts = drift or differentiation.

Same process, applied to yourself. Fetch your own sitemap. Extract your blog and evergreen page URLs. Group them into the same topical buckets you built in step 2.

Step 4: Diff the topic clusters

Bucket-level first (which buckets have 5+ competitor URLs and 0-2 yours?). Then URL-level within each. Dedupe synonyms.

You now have two topical maps: theirs (5 competitors × ~8 buckets) and yours (1 site × 8 buckets). Diff them at the bucket level first, then at the URL level.

1. 1Bucket-level diff: which buckets have >5 competitor URLs and 0-2 yours? Those are your priority buckets.
2. 2URL-level diff: within each priority bucket, list the specific competitor URLs. Look for repeated topics. If 3 of 5 competitors have a "redirect chain checker" post, that topic has confirmed search demand. If only 1 competitor has it, treat it as speculative.
3. 3Deduplicate synonyms. "Bulk redirect checker" and "redirect chain audit tool" are the same buyer intent. Merge them.

The output of step 4 is a raw target list, usually 15-25 topics per 5 competitors. That's still too many to write against. Step 5 prioritizes.

Step 5: Score gaps by volume × rank × authority

Score each target on 3 dimensions (1-5), multiply, sort. Top 5 = your immediate draft queue.

Rank each target topic on three dimensions. Score 1-5 on each. Multiply.

1. 1Search volume signal. Free proxies: number of competitors ranking (3+ = high demand), Google "People Also Ask" presence (yes = high), Perplexity giving a substantive answer (yes = high). Do not use Google Keyword Planner without a paid Ads account, it withholds real numbers from free users.
2. 2Competitor rank strength. If the top-ranking competitor URL is on page 1, score low, that URL has authority you're not overtaking easily. If competitors are on page 2-3, score high, those pages are beatable. Check with a private-window search from a cold browser session.
3. 3Your existing authority. Do you already have any URL ranking anywhere in the top 100? If yes, score 5, an existing weak-ranking URL is much cheaper to upgrade than a new URL is to launch. If nothing exists, score 3. Never touched the topic, score 1.

Multiply the three. Sort descending. Your top 5 targets are the ones you draft first. Everything below rank 10 goes into a backlog.

The manual version: spreadsheet + Ahrefs free tools

45 minutes for 5 competitors. Zero paid subscriptions. Misses topics not in sitemaps + topical drift.

The full manual playbook, timed on a fresh workspace:

1. 1Minute 0-5: list 5 competitors in column A.
2. 2Minutes 5-15: fetch each sitemap, filter to blog URLs, paste into columns B-F.
3. 3Minutes 15-25: read the slugs, eyeball-cluster into topic buckets in column G.
4. 4Minutes 25-30: repeat for your own site in a second sheet.
5. 5Minutes 30-35: diff at the bucket level, list target topics in column H.
6. 6Minutes 35-45: score each topic on the 3-dimension rubric using Ahrefs' free traffic checker and Google Trends for volume proxies, private-window Google searches for rank checks.

The CiteClip version: 8 minutes, continuous

Auto-discovers competitors, crawls sitemap + RSS, embeds + clusters with text-embedding-3-large, re-scores every ~41h.

Same playbook, automated. During onboarding, CiteClip asks for your domain and auto-discovers your 5 direct competitors from public data (usually 90 seconds). It then crawls each competitor's sitemap and RSS feed, extracts the last 50 blog posts, embeds each with text-embedding-3-large, and clusters them into topic buckets automatically.

Your own site gets the same treatment. The diff runs continuously, every time a competitor publishes a new post, the topic map updates and the gap re-scores. Search volume, rank strength, and authority scores are calculated from live Google + Perplexity signals.

When automation pays off, and what to do this week

1-2 posts/month: manual is fine. 4+ posts/month: automate. Continuous freshness compounds week over week.

This week: pick your 5 direct competitors right now. If you can't name them off the top of your head, you're not ready to do gap analysis, go do positioning work first (the five-page framework post covers what to publish once you know your competitors). If you can name them, run the manual playbook in 45 minutes and you'll have a target list by end of day.

If you'd rather have the automated version (continuous monitoring, embedding-based clustering, AI-drafted articles against each identified gap) sign up at citeclip.com. First 4 articles are free, no credit card required. Compare the output against your gut. If the ranked list matches what you'd have picked manually, keep drafting. If it doesn't, email us, that's the exact feedback loop we need.