Products Solutions Templates Pricing Resources

Configuration best practices: settings that prevent data quality issues

Home / Everything About / Everything About Analytics / Configuration best practices: settings that prevent data quality issues

Most data quality problems start with bad configuration. A filter that's too broad. Tracking tags on the wrong pages. Conversion goals defined incorrectly. These aren't tracking failures. They're setup mistakes that silently distort data.

Good configuration prevents most data quality issues. The right setup from the start avoids months of wondering why your data doesn't make sense.

Configuration best practices

Start with clear goals

Before setting up analytics, define what you want to measure. What are your conversions? Which pages matter? What traffic sources matter most? Clear goals prevent over-complicated setups.

Document everything

Document why each filter exists. Document how each conversion is defined. Document why you chose a specific session timeout. Documentation prevents someone from accidentally breaking working configuration.

Use separate views for experiments

Create a test view for new filters, goals, or configurations. Test there first. Only apply to production view once tested.

Never filter without testing

Create a filter on a test view first. Watch it for a week. Make sure it excludes only what you intend to exclude. Then apply to production.

Keep production configuration simple

Simple configuration is less likely to break. Complex multi-step filters, many custom metrics, complicated goals—all add failure points. Start simple. Add only what you actually need.

Specific configuration best practices

Set timezone once and forget it

Pick your business's primary location. Set timezone to that. Never change it. Create a new view if you need a different timezone for comparison.

Use standard session timeout

30 minutes is the industry standard. Unless you have a specific reason to change it, leave it at 30. Changing it breaks historical comparisons.

Create internal traffic filter

Filter out traffic from your office, your team, your testing. Use an IP-based filter. Test on a secondary view to make sure you're excluding only your traffic.

Enable bot filtering

Turn on "Exclude hits from known bots and spiders." This removes obvious bot traffic without false positives.

Define conversion goals clearly

For each goal, write down exactly when it triggers. "Goal triggers when user reaches /thank-you page." Not "goal triggers on purchase" (which is ambiguous).

Use consistent namings

Consistency prevents mistakes. Use the same naming convention for all events. If you call one event "purchase" and another "order," you create confusion.

Regular configuration audits

Quarterly, review your configuration. Are all filters still needed? Are all goals still accurate? Has anything changed? Remove dead configuration.

Avoiding common configuration mistakes

Overly broad filters

A filter meant to exclude your office IP accidentally excludes your customer's office too. Test filters carefully. Start narrow. Expand only if needed.

Forgetting to implement tracking on new pages

Launch a new section of your site without tracking code. Suddenly that section is invisible. Implement tracking on every new page during dev, not after launch.

Changing goals retroactively

If you change how a goal is defined, historical data doesn't change. Old data uses old definition. New data uses new definition. They're incomparable. If you must change a goal, create a new one instead.

Using complex regular expressions

Regex filters are powerful but easy to break. A small mistake excludes things you didn't intend. Use simple filters when possible. Test complex ones thoroughly.