In this guide
- Check accuracy, completeness, consistency, timeliness and uniqueness for your key events.
- Fix problems at the source (data layer, forms), not downstream.
- Track a scorecard and name an owner.
"Our data is bad" is a feeling. Data quality work turns it into numbers you can improve. This page defines five properties, shows how to test each for tracking data and gives a scorecard. It extends data validation techniques and monitoring website tracking.
Five properties
| Property | Question | Example test |
|---|---|---|
| Accuracy | Is the value right? | Purchase value equals the order total |
| Completeness | Is anything missing? | Share of purchases with currency and transaction ID |
| Consistency | Do systems agree? | Platform purchases against store orders |
| Timeliness | Does it arrive when it should? | Median delay between order and event |
| Uniqueness | Is anything duplicated? | Duplicate transaction IDs |
Watch events arrive, live
A sample feed. In Open Tracked you see the same stream for your own store.
Controls at the source
- A written event specification: fields, types and rules. See tagging strategy.
- Validation in the data layer code: required fields, number formats.
- Form validation: formats and consent flags.
- Event IDs generated once and shared.
Controls in the container
- Triggers that skip events with missing required values, or flag them.
- Transformations that normalise values (currency, lower-case e-mail). See transformations.
- Hashing done once, in one place.
- Open Tracked's tracking score as a quick set-up check: server running, own subdomain, CDN, Custom Loader, Cookie Keeper and events arriving.
Brands that grow on cleaner data
Stores and agencies use Open Tracked to see which ads earn the sale.
Monitor
Add a daily or weekly check of the scorecard. Alert when a measure crosses a threshold. See monitoring website tracking.
A scorecard
| Measure | Target | Actual |
|---|---|---|
| Purchases with transaction ID and currency | 100 percent | |
| Duplicate transaction IDs | Under 0.5 percent | |
| Platform purchases รท store orders | Stable, above your baseline | |
| Median event delay | Under a minute | |
| Leads with consent recorded | 100 percent |
Ownership
Name one person for each event and each platform connection. Data quality improves when someone is responsible, and decays when everyone is.
A realistic example
A store notices that its dashboard revenue is 12 percent higher than its accounts. It builds a small scorecard. Completeness is fine: every purchase has an ID, value and currency. Uniqueness is the problem: 6 percent of transaction IDs appear twice, because the thank-you page re-fires when customers refresh. Consistency shows platforms seeing slightly more purchases than the shop, which fits. The fix is a once-per-order guard on the thank-you page. Within a week duplicates fall under 0.5 percent, dashboard revenue matches the accounts within a point, and the team writes the scorecard into its weekly routine. The method was simple. The value came from measuring each property separately instead of saying the data was bad.
Keep the scorecard small. Five to eight measures that you review every week are worth more than fifty that nobody opens. When a measure goes red, trace it to a source: a form, a data layer rule, a tag. Fix the source, then check that the measure recovers.
Finally, write down what you decided to measure and why, so a successor can continue. Data quality work often stops when the person who started it moves on. A short note with the scorecard, the thresholds and the sources of each measure keeps the practice alive. Review the thresholds once a year, since your volume and your tools will change, and a target that was sensible last year can be too loose or too strict today.
Connected in minutes
Add your domain, install the pixel, connect your ad platforms and watch the first events land.
Frequently asked questions
What is data quality in tracking?
How accurate, complete, consistent, timely and unique your tracking data is.
Where should I fix data quality?
At the source: the data layer, forms and event rules.
How do I measure it?
With a scorecard of tests for each property.
How often should I check?
Weekly for key events.
Who should own it?
A named person for each event and platform connection.
Do I need spreadsheets to read my results?
No. Once tracking is live, the Open Tracked analytics shows spend, revenue, ROAS and profit, and you can ask Ari, the built-in AI assistant, questions about your own numbers in plain words. Ari's allowance depends on your analytics plan.
Is Open Tracked different from Open Track or Open Space Track?
No. Open Tracked is the official name of our server-side tracking platform, and the website is opentracked.io. Many people search for it as Open Track, OpenTrack or Open Space Track. All of them lead to the same product: server-side Google Tag Manager hosting, first-party analytics and a CRM pipeline.
Fix your tracking at the source
Open Tracked hosts your server-side Tag Manager, keeps first-party cookies alive and shows what each ad is really worth.
The Open Tracked team
We build Open Tracked: server-side tracking hosting, analytics and the tools around them, for stores that spend on ads. Questions about this guide, or an idea for the next one?
Comments
Loading comments...
Integrations · 5 min read
