Analytics & tracking setup checklist

You can't improve what you don't measure—but tracking setup is also one of the most common things skipped until after launch, which means you lose your baseline data for the exact period you'll want it most: launch week.

Choosing what to measure

• Metrics tracked actually connect to the site's real goals—leads, sales, sign-ups, support deflection—rather than defaulting to whatever a dashboard shows first
• Vanity metrics (raw pageviews and visits alone) are treated as context, not as the primary measure of whether the site is succeeding
• A small set of key metrics is agreed on in advance, so success or failure isn't judged retroactively by whichever number happens to look good
• Metrics are tied to specific pages or flows where relevant, so a problem can be traced to where it's actually happening rather than only seen in an aggregate number

Before you launch

• An analytics tool is installed and verified as firing correctly before launch day, not added afterward
• Google Search Console is set up and the site's ownership is verified
• The XML sitemap is submitted inside Search Console once the live domain is verified
• Tracking has been tested on a staging environment to confirm it doesn't pollute production data once live

Core setup

• Analytics tracking code is installed on every page, not just a subset
• A tag management system is used (if your setup benefits from one) so tracking tags can be managed without editing site code directly
• Internal traffic (your own team's visits) is filtered out or excluded so it doesn't skew reports
• Cross-domain tracking is configured correctly if your site spans multiple domains or subdomains

Goals & conversions

• Key actions—form submissions, purchases, sign-ups, calls—are set up as trackable conversions or events
• Conversion tracking has been tested by completing the action yourself and confirming it registers
• E-commerce tracking is configured correctly if the site sells products (see the e-commerce checklist)

Attribution & campaigns

• Marketing links (email, social, ads) use consistent, well-structured tracking parameters so traffic sources can actually be distinguished later
• Attribution settings are understood well enough to interpret reports correctly, rather than treating a single attribution model's numbers as unambiguous truth
• Referral traffic from payment processors or embedded third-party checkouts isn't accidentally misattributed as a normal external referral

Data quality

• Reported numbers are spot-checked against a real, known event—a test purchase, a known form submission—to confirm tracking actually reflects reality
• Bot and spam traffic is reviewed for and filtered where the analytics platform supports it
• Sudden, unexplained jumps or drops in a metric are investigated rather than reported at face value

Privacy & compliance

• A cookie/consent mechanism is in place if required in your audience's jurisdiction, and analytics respects the visitor's choice (see the privacy & cookie consent checklist)
• The privacy policy accurately describes what's actually being tracked and why
• Data is not collected beyond what you actually need and can justify
• IP anonymization or equivalent privacy-preserving settings are enabled where your analytics tool supports it

Reporting to stakeholders

• Reports are kept simple enough that the people who need the data actually read and use it, rather than a dashboard nobody opens
• The metrics being reported are the ones that actually matter to the business's goals, not just whatever's easiest to pull
• Context is included alongside numbers—a metric without an explanation of what changed and why is rarely actionable

Ongoing use

• A regular cadence is set for actually reviewing analytics—collecting data nobody looks at isn't useful
• Search Console's performance, coverage and Core Web Vitals reports are checked periodically, not just at setup
• Reports are cross-checked occasionally against a second source to catch tracking that's silently broken
• Historical data is preserved when switching or upgrading analytics platforms, rather than accepting a clean break that erases the ability to compare against past performance
• Segment- or channel-level performance (which traffic sources, which pages) is reviewed periodically, not just the site-wide total, since an overall average can hide a channel that's quietly underperforming

Pair analytics setup with the technical SEO checklist for Search Console fundamentals, the conversion & forms checklist for what to actually do with the data once you have it, and revisit it as part of post-launch maintenance.