Loading tutorials…
Loading tutorials…
A SEMrush Project is the container that pipes data into Site Audit, Position Tracking, On-Page SEO, Listing Management, and Social Tracker. Configure it wrong on day one and every downstream module produces noise for months. This is the right setup.
Who this is forOwners or in-house marketers paying $139-$249/mo for a SEMrush Pro or Guru plan who haven't actually configured a Project yet — or who configured one in a hurry and want to redo it cleanly. If your Site Audit returned 3,000 issues and you ignored them, that's the signal.
What you'll need
Step 1
SEMrush → Projects (left rail) → Add new project → Domain → name. Match the project domain to your canonical host.
Open SEMrush, click Projects in the left rail, then Add new project. Enter the root domain — example.com if your canonical is apex, www.example.com if you 301 to www.
The domain you enter here becomes the scope for Site Audit, On-Page SEO Checker, and Backlink Audit. Get it wrong and every report you pull is technically reporting on a redirect target — credit waste plus distorted metrics.
Give the project a name a teammate will recognize 6 months from now. 'example.com — production' beats 'My Project (2).'
Click Create project. SEMrush opens the project dashboard with 14 tool tiles, each of which needs configuration. Don't enable all 14 at once — we'll walk the priority order.
Step 2
Project dashboard → Connect integrations → Google Search Console + Google Analytics. This unlocks real query data in Position Tracking and Site Audit.
On the project dashboard, find the Integrations section (top right, or under settings). Click Connect for Google Search Console.
Authenticate with the same Google account that owns the GSC property. Pick the matching property (apex vs www must match the Project domain).
Repeat the connection for Google Analytics 4. SEMrush will ask for read access on the GA4 property and a specific data stream.
GSC integration unlocks: real query data in Organic Research, click data in Position Tracking, and validated indexed-URL counts in Site Audit. Without it, you're reading SEMrush's modeled estimates instead of your real data.
Verify the connection by opening Position Tracking → if GSC is connected, you'll see a 'Connected' badge in the integrations panel.
Step 3
Site Audit tile → Settings → Crawl source, JS rendering, URL limits, exclusion rules. JS rendering doubles credit cost — only enable when needed.
Click the Site Audit tile on the project dashboard. If first-time, the setup wizard opens. If already configured, click the gear icon → Site Audit Settings.
Crawl source: Website is the default, but enable Sitemaps on site AND URLs from GSC for full coverage. The combination catches orphan-indexed pages that a pure crawl misses.
JS rendering: enable ONLY if your stack is client-side rendered (CSR React, Vue without SSR, plain Next.js without getServerSideProps). Check by opening View Source on 3-5 pages — if titles, body, and meta description are in the HTML, leave JS rendering OFF.
URL limit: Pro plan caps at 100K URLs/month; Guru at 300K; Business at 1M. If your site has 2,000 URLs, leave the limit at the plan default — no need to throttle.
Exclusion rules: under Crawler settings → Allow/Disallow URLs, exclude faceted-nav (?filter=, ?sort=), search results (/?s=, /search/), and session-ID parameters. Faceted nav is the #1 crawl-credit waster on ecom sites.
Crawl speed: SEMrush default is 'Minimum delay between pages.' If your server is healthy, this is fine. If your CDN is throttling, switch to '1 URL per 2 seconds' as a safety valve.
Step 4
Project dashboard → Position Tracking tile → Setup. Target location, device, search engine, then upload 20-100 priority keywords.
Click the Position Tracking tile and run the setup wizard. SEMrush asks for: target location (city/state/country), device (Desktop, Mobile, or both), search engine (almost always Google), and language.
Pick the most specific location your real customers search from. For local-intent SEO, drop down to city level. For B2B/national, country-level is fine.
On the keyword input step, paste 20-100 priority keywords. Don't dump 500 — the Pro plan caps at 500 tracked keywords total and you want headroom for new clusters.
Add 2-5 competitors in the Competitors field. SEMrush will track your rankings alongside theirs in the same SERP context.
Set up tags (e.g., 'brand,' 'category-page,' 'BoFu,' 'cluster-X') so you can filter the dashboard meaningfully later. Untagged keywords become a 500-row list nobody reads.
Step 5
Project dashboard → On-Page SEO Checker tile → Setup, then Backlink Audit tile → Setup. Both pull from your tracked keywords and connected GSC.
Click the On-Page SEO Checker tile. The wizard asks for: target location (match Position Tracking), keywords (import from Position Tracking with one click), and target URLs.
For target URLs, let SEMrush pull from GSC's top landing pages automatically. Manually override only if you have a specific cluster you want to focus on.
Click the Backlink Audit tile. The wizard asks for: target country (where your business operates), target categories (industry tags — pick 2-3 that match your business), and competitors (3-5).
Backlink Audit will pull your backlink profile from SEMrush's index, assign toxicity scores (0-100), and flag patterns that look algorithmic-penalty-prone.
Both tools take 24-48 hours to populate after first setup. Don't read the data until that initial pull completes.
Step 6
Start the first Site Audit crawl. Compare crawled URL count against GSC Pages → Indexed once complete. Mismatch over 25% means scope or sitemap is wrong.
Open Site Audit → Start Audit (or wait for the scheduled crawl). For sites under 5K URLs, the crawl finishes in 30-90 minutes. For 50K+ URLs, plan 4-12 hours.
While it runs, open Google Search Console → Pages → Indexed count. Note the number.
Once SEMrush's Site Audit completes, compare crawled URLs in Site Audit Overview vs GSC indexed count. If SEMrush is off by more than 25% in either direction, your scope or sitemap is wrong. Diagnose before acting on the data.
Spot-check 10 random URLs from SEMrush's crawl report against the live site. If SEMrush is showing 'noindex' on a page that's actually indexable in the browser, JS rendering may be OFF when it should be ON.
Site Health score appears in the top-left after the crawl finishes. Target: 85+ for content sites, 90+ for ecom. We'll triage the underlying issues in the next tutorial.
Step 7
Site Audit Settings → Schedule → Weekly. Position Tracking is auto-daily. Set up email digests so the data stays in front of the team.
Site Audit → Settings → Schedule → Weekly. Pick a day when nobody on the team is deploying (Tuesday or Wednesday usually).
Position Tracking auto-refreshes daily — no schedule needed.
Set up email digests under Project Settings → Email Reports. Configure a weekly email summary of: new Site Audit errors, Position Tracking movement (gainers + losers), and new toxic backlinks flagged by Backlink Audit.
Send the digest to a single owner (not a distribution list). Distribution lists get ignored. One owner with accountability gets read.
Add a calendar reminder for monthly Project review — 30 min/month is enough to catch drift before it compounds.
Common mistakes
Enabling JavaScript rendering "just in case"
What goes wrong: Crawl credits drain at 2x the rate. A Pro plan that should cover 100K URLs/month suddenly covers 50K. Hit the cap mid-month and Site Audit stops auditing your most-important pages. You waste roughly $70/mo in unusable capacity.
How to avoid: Check View Source on 5 representative pages. If content (title, body, meta description, schema) is in the HTML, leave JS rendering OFF. Re-enable only if a spot-check shows content missing without it.
Crawling at full scope without exclusion rules
What goes wrong: Site Audit returns 40,000 URLs when your real site is 4,000. Faceted-nav parameters dominate the report. You waste 8 hours filtering noise that should never have been crawled — that's $1,000-2,000 of time at typical founder hourly value.
How to avoid: Add URL exclusion rules under Site Audit Settings → Crawler settings → Disallow URLs: exclude ?s=, ?filter=, /search/, ?page=, session-ID patterns. Re-run the crawl with the cleaner scope.
Mixing apex and www domains across modules
What goes wrong: Project is example.com but Position Tracking is configured for www.example.com. The two modules track different URL universes. Your rankings dashboard and your audit don't reconcile. 6 months of contradictory reports.
How to avoid: Pick ONE canonical host (whichever your 301 resolves to). Use that exact form in every Project module. If you discover a mismatch, delete and rebuild the misconfigured module — patching it half-fixes the data.
Not connecting Google Search Console
What goes wrong: Position Tracking shows modeled estimates instead of real GSC click data. Organic Research is missing your zero-click and brand queries. Site Audit can't surface URLs that are indexed but not in your sitemap. You pay for $250/mo of features and use 60% of the value.
How to avoid: Project dashboard → Integrations → Google Search Console → Connect. Use a Google account that has Owner or Full User access to the GSC property. Verify the 'Connected' badge appears in the integrations panel.
Tracking 500 keywords from day one
What goes wrong: Pro plan caps at 500 tracked keywords. Burning all 500 on initial setup leaves no room for new clusters as your content grows. Six months later you're juggling which old keywords to delete to make room.
How to avoid: Start with 50-100 high-priority keywords (brand, top-of-funnel BoFu queries, money pages). Add 20-50/month as new content ships. Stay below 80% of cap so you have room to test new clusters.
Treating Site Health Score as the goal
What goes wrong: Site Health Score is a weighted average. Chasing 100 means fixing low-impact issues (missing alt text on decorative images) while leaving high-impact problems (broken canonicals, 5xx on money pages) unaddressed. You burn 40 hours and the score moves 3 points.
How to avoid: Use Site Health as a trend indicator only. The real metric is 'Errors' count and 'URLs affected.' Tackle top 10 errors by URLs affected, not by alphabetical issue list. Move the trend, not the headline number.
Recap
Done — what's next
SEMrush keyword research workflow with Keyword Magic Tool
Read the next tutorial
Hand it off
Configuring a SEMrush Project once is a 2-3 hour task. Running the audits weekly, triaging issues, validating Position Tracking movement, and shipping the fixes is a job. A vetted technical SEO specialist on EverestX will own the Project + the cadence + the fixes — typically $400-1,000/mo at $14-16/hr depending on site size.
See specialist rates
Pro ($139/mo) gives 5 projects, 500 tracked keywords, 100K Site Audit URLs/month — enough for a single brand. Guru ($249/mo) gives 15 projects, 1,500 keywords, 300K URLs, plus Content Marketing Toolkit and historical data back to 2012. Pick Guru if you need historical trends, run multiple brands, or want Content Marketing Toolkit.
Open 3-5 representative pages. Right-click → View Source. If body copy, title, and meta description are in the HTML, you don't need JS rendering. If those fields are empty placeholders that get filled by React/Vue/Next client-side, you do. Enabling it doubles credit cost — verify before turning it on.
Three usual reasons: (1) GSC integration isn't connected, so SEMrush only sees URLs it discovered via crawl; (2) SEMrush is crawling staging/parameter URLs that aren't in GSC's index; (3) your sitemap is incomplete and SEMrush is following different paths. Connect GSC and add exclusion rules to reconcile.
First check whether the crawl scope changed (new sitemap entries, new exclusion rules removed, a deploy added 200 pages). Score drops are almost always explained by scope changes, not real degradation. Look at 'Errors' count delta — if that's flat but score dropped, it's a scope artifact, not a problem.
Wait 7-14 days after fixes ship before re-crawling. CDN caches, crawl-discovery delays, and CMS deploy latency mean immediate re-crawls show stale state. Weekly schedule is enough for most teams.
No — SEMrush Site Audit requires domain verification (or implicit verification via the project setup). For competitor analysis, use Organic Research or the Backlink Analytics module, both of which work on any domain without verification.
SEMrush
Keyword Magic Tool returns 30,000 keywords from a single seed. Most teams export the first 200, dump them into a spreadsheet, and never ship a piece of content from the list. This is the workflow specialists actually use — from seed to ranked content brief in under 5 hours.
SEMrush
Site Audit returns 3,000 issues. You have time to fix 30. Picking the wrong 30 wastes a quarter; picking the right 30 lifts rankings within 60 days. This is the priority order specialists use.
SEMrush
Position Tracking is SEMrush's most-used module — and the most misconfigured. Wrong location, mixed device targets, untagged keywords, and missing competitors mean the dashboard becomes noise. This is the right configuration.
SEMrush
You're paying $139-$500/mo for SEMrush. The question isn't whether the tool is worth it — it's whether you have the time and skill to actually use what you're paying for. This is the honest framework.