Read the Proof Impact report
Read Proof Impact's totals, attribution models and breakdowns, check data quality, fix a value that shows Unavailable and export the numbers.
The Proof Impact report shows how many of your conversions followed a widget view, and which widgets, testimonials, Proof Match rules and pages they followed. It's on Premium and Business, and anyone who can open the project can read it. The workspace owner and the project's Admins can also export it and change the conversion events.
The report appears once your project's first conversion has arrived. Until then the Overview tab shows a setup checklist: see Set up Proof Impact tracking.
Open the report and pick a date range
Open Proof Impact under Analyze in your project's sidebar.
Pick a date range at the top: 7 days, 30 days, 90 days (the default) or All retained, or open Custom range. Dates are in UTC, and today stays incomplete until it's over. The range and the filters you choose stay on when you switch tabs.
Read the totals on the Overview tab
The Overview tab shows the conversion event chosen under Reporting scope (your primary event unless you pick another, or All included events (mixed outcomes)). Five figures sum it up, from Total included conversions to Avg value / attributed: see What each figure means. Below them, Impact over time charts conversions, attributed conversions and value by day.
Above the figures, Tracking health counts widget loads, widget views and conversions from the last 24 hours; View diagnostics opens the details. When something needs attention, a panel names it, such as Choose your primary conversion or No conversions in this reporting range. For what to do about each panel, see Proof Impact shows no data or unmatched conversions.
Break the numbers down on the Attribution tab
The Attribution tab has the same five figures and chart, and the detail behind them:
- The model buttons at the top decide which widget view gets the credit when a visitor saw several: Last-touch (the default), First-touch or Fractional. See Attribution models.
- Filters narrows the report by Widget, Page, Contextual rule, Testimonial, Conversion event and Currency.
- Top widgets, Top testimonials, Top contextual rules and Top page paths list widget views, attributed conversions and value for each. Click a rule or a page in its list to filter the report by it.
Check the Data Quality tab
The Data Quality tab shows how complete the numbers are:
- Unmatched conversions lists the ten latest conversions that no widget view matched, with the reason for each. For what each reason means, see Conversions arrive but aren't matched.
- Data Health has six percentages: see Data Health figures.
- How long data is kept gives the number of days your project keeps detailed widget views, conversions with their attribution, and the daily report history.
- Tracking configuration sums up the project's tracking settings. See Privacy.
- Conversion events holds your events, each with its conversions in the date range. See Change an event later.
Export the numbers
The workspace owner and Admins can download the report from Export, at the top of the Overview, Attribution and Data Quality tabs. Each file follows the date range and the filters on screen. The conversion ledger leaves out the Page, Contextual rule and Testimonial filters, which describe widget views.
The export files says what each file holds and how many rows it takes.
The export files
Export offers three files:
- Attribution summary CSV: the totals and the four Top lists.
- Conversion ledger CSV (first 1,000 rows): one row for each conversion.
- Attribution touch ledger CSV (first 1,000 rows): one row for each credit a conversion gave to a widget view and a testimonial.
A ledger file holds up to 1,000 rows. When the range has more, the rows at the top of the file say so beside Has more rows and give a Next page URL. Open that address in the browser where you're signed in to download the next file.
What each figure means
- Total included conversions: every conversion of the included events in the range.
- Attributed conversions: conversions that had a qualifying widget view inside the event's attribution window.
- Attribution coverage: attributed conversions divided by all included conversions.
- Attributed value: the conversion values credited to widget views, in one currency. A badge says whether the amounts came from visitors' browsers (Browser-reported), your server (Server-authenticated) or both (Mixed trust).
- Avg value / attributed: attributed value divided by attributed conversions.
While you filter by a widget, page, rule or testimonial, the first two tiles become Conversions with scoped attribution and Unique scoped conversions, and Attribution coverage shows a dash.
Attribution models
The model decides which widget view gets the credit when a visitor saw several: Last-touch (the default) credits the last one, First-touch the first, and Fractional splits the credit equally among all of them. Each view's share is then split among the testimonials it showed, so the counts in the Top lists can have decimals.
Data Health figures
Each figure covers the date range. The four about conversions also follow the event chosen under Reporting scope. On the card, each figure has a line that says what it counts, and whether higher or lower is better.
| Figure | What it is |
|---|---|
| Widget view session coverage | The share of widget views that arrived with the ID of the visitor's browser tab. A view needs that ID to be matched to a conversion in the same tab. |
| Conversion session coverage | The share of conversions that arrived with that ID. One without it is listed under Unmatched conversions as Missing session ID on conversion. |
| Unattributed conversions | The share of conversions that no widget view matched. |
| Missing value | The share of conversions that came without an amount. |
| Matched from the browser's record | Of the attributed conversions, the share credited to a view known only from the list of recent views that the visitor's browser sent with the conversion. |
| Proof Match rule coverage | Of the views of your Proof Match widgets, the share where one of the widget's rules applied. A dash means no Proof Match widget was viewed. |
If value shows "Unavailable"
Attributed value needs amounts in a single currency. It shows Unavailable, and so does Avg value / attributed, with a line under it that says which of two cases applies:
- Your conversions use several currencies. A notice, Attributed value is grouped by currency, names them. Pick one in the Currency filter of the Attribution tab; the filter stays on when you switch tabs. Amounts in different currencies are never added together.
- No conversion in view came with an amount, for example because the event's Value policy is No value expected. To see value, give the event a policy that takes one and send the amount and its currency with each conversion: see Event settings.
Change an event later
In Data Quality under Conversion events, click Edit to change an event's Display label, Outcome role, Value policy, Attribution window, Identity scope or Reporting switch, then Save changes. A new window or identity scope applies to new conversions only; past results aren't recalculated. An event whose Reporting is off stays out of the totals. Archive refuses new conversions for the event, and its past conversions leave the report's totals, chart and lists too. Reactivate brings the event and its history back.
An event can't be deleted, and its Event key can't be changed. An archived event keeps its key, so a new event needs another one.
Only one active event can be the Primary outcome. To make another event primary, give the current one another role first. On a widget that uses Auto-Optimize, a new primary event starts its tests again.
For what each setting does, see Event settings.
What happens next
The report tells you what visitors saw before they converted, not whether the widget made them convert. To find that out, run an experiment.
Was this page helpful?
Set up Proof Impact tracking
Create a conversion event, add the tracking code to your site and send a test, so Proof Impact can link conversions to widget views.
Run a Proof Impact experiment with a no-widget group
Split your visitors between a published widget and no widget at all, then compare how often each group converts.