What you will learn #
Use form analytics to see where sampled visitors start a form, attempt to submit and leave. Investigate forms and fields without treating an attempt as server-confirmed success. The aggregate report does not show the answers people typed.
Before you begin #
Availability: This feature is being opened gradually. An eligible plan or trial does not by itself enable it for every website. The following instructions apply when your website has access; follow its displayed access message.
- Add Hitsteps to the site and confirm that Experience is collecting recordings for the pages you want to measure. Form analytics is based on retained, sampled Experience visits, so it is not a census of every visitor or a replacement for your form processor's submission records.
- Form tracking must be available for your website and enabled in its Experience settings. Access is currently gated by the Experience service and account/site access. Check the access screen for your website and the current features and plans. A plan listing alone does not guarantee that the feature has been opened for your account.
- You need dashboard read access to view the report. Editing form display names and changing collection settings require website-management access. Ask the website owner if you cannot see those controls.
- Use a representative date range and page. The report covers retained visits and can show partial current-day data. It is designed for aggregate analysis, not to retrieve a visitor's form entry.
Step by step #
Read a form report
The report groups forms by page path and form identity. Start with the completion summary, then select a form to inspect its fields and related recordings.
- Choose the date range, page path, and device that match the question you are investigating. Custom ranges are inclusive, in UTC, and limited to 90 days. The 7, 30, and 90 day shortcuts use completed UTC days through yesterday.
- In the form list, compare forms on the same page or open one form for its detail view. A form's detected name can be replaced with a safe display name if you have permission.
- Review the counts and rates together. A count of starts is the denominator for completion rates. A missing success signal is shown as unknown, not as a zero success rate.
- Inspect field patterns, validation signals, repeat focus, and last-field exits. These are clues for investigation, not explanations of visitor intent. Follow a report link to the associated retained recording when one is available and you have access.
- Compare devices and daily trends only when each selection has enough coverage. Treat changes as correlations; device mix, sampling, consent, blockers, and retention can change the observed sample.
- Use the page's Review next patterns to decide what to inspect. A validation error or repeated focus says where a problem may be happening, not why it happened. Check the form on your own test session before changing the live form.
Illustrative example
Suppose the report shows 40 started visits and 24 submitted visits for a contact form in the selected sample. The submission rate is 24 ÷ 40 = 60%. If the report also shows that a small number of retained visits reached the address field and then ended, treat that as a prompt to test the form on that page and device. These figures are illustrative only; they are not Hitsteps customer results or a benchmark.
Set up collection and readable names
Check the two collection controls
- Open Experience → Settings for the selected website.
- Find Form tracking and Form interactions. The first enables the form analytics report; the second allows form interactions to be captured as part of Experience recordings. The report requires both controls and the service-side availability gate.
- Read the page's privacy and masking settings before changing either control. Form interactions can include typed text in session replay according to the Text privacy setting; password and payment fields remain masked. Form analytics is a separate aggregate view and does not include field answers.
- Save only if you are authorized and have decided that collection fits your site's privacy notice, consent choices, and applicable requirements. Return to Experience → Form tracking and check for new sampled visits after the change. A settings change does not recreate past data.
If your account has a heatmap-only collection mode, form analytics is disabled in that mode. Do not switch modes just to obtain form data; use the normal product access and owner-approved settings path.
Use stable, non-personal names (optional)
Forms and fields are detected automatically. If an application re-renders them and the detected identity is hard to recognize, a developer can add static labels to the form markup:
<form data-hitsteps-form="contact">
<label for="contact-email">Email</label>
<input id="contact-email" name="email" type="email" data-hitsteps-field="email">
</form>
Use short, stable labels such as contact and email. Never put a visitor's name, email address, account number, or other personal information in these attributes. The labels identify report rows; they are not a place to send or store form answers.
If you have website-management access, use Edit form name or Edit field name in the report to set a display name. Enter a plain-text, non-personal label and choose Save name. Use Use detected name to remove the override. These names are report labels only. A saved override applies to the form identity and page shown in the report.
Optional outcome signals for application forms
For applications that submit in the background without a page navigation, a developer may use the Hitsteps Experience JavaScript outcome methods to associate the application's actual result with an attempt. Call formAttempt(formElement) before the application's normal submission; after the application resolves, call formSuccess(formElement, attempt) or formFailure(formElement, attempt). Keep the normal application request running even when the tracking object is absent. This optional integration calibrates form analytics; it is not required to count a native submit attempt. Use the current form tracking explanation and have a developer verify the API against the site's current integration before adding code.
Do not treat an automatic success estimate as proof that a form was accepted by your server. For important leads, orders, or account changes, use the system that processes the form as the authoritative record.
Worked example #
Illustrative example: A form has 40 started visits and 24 submitted visits in the selected sample. Submission rate is 24 ÷ 40 = 60%. Last-field exits suggest a place to test; they do not establish the cause or represent every visitor.
Controls and metrics #
| Measure | Meaning in the report | How to use it |
|---|---|---|
| Viewed | A retained sampled visit in which the form was observed. | Use it to understand the sample behind the form report. |
| Started | A viewed form visit with a tracked interaction. | This is the denominator for submission, success, and abandonment rates. |
| Submitted | A native submit or application-reported submission attempt. Multiple attempts in one visit can appear in attempt-event counts. | It means an attempt was observed, not that your server accepted the business transaction. |
| Success | Explicit application confirmation or a clearly labelled automatic estimate when the browser can attribute a completed navigation to that form. | An automatic estimate is not server-side order, lead, or account confirmation. AJAX and same-page flows may need an optional application outcome signal. |
| Left without submit / Abandoned | A started form visit that ended before a submission was observed. | This is an observed outcome in the retained sample, not proof that a person intended to submit or that the form caused the exit. |
| Submission rate | Submitted form visits divided by started visits. | Compare like-for-like pages, ranges, and devices. |
| Success rate | Effective successes divided by started visits, including automatic estimates where shown. | If the report says Signal not configured, success is unknown for that selection; do not read the dash as 0%. |
| Validation affected | A form/field pattern involving a browser validation block. | Inspect the form and its error message in a safe test session. |
| Repeat focus | Repeated attention to a field in the observed sample. | It can suggest uncertainty or correction, but does not identify the cause. |
| Last-field exit | An observed form visit that ended after the last tracked field. | Treat it as a location clue, not a diagnosis. |
| First interaction → submission | Average measured time from first interaction until a native submit or explicit attempt, excluding validation-blocked attempts. | Only visits with the required timing signal are included. |
| Measured outcome wait | Average time from an attempt to an explicit success or failure outcome. | Only measured outcomes are included; this is not a server response-time metric. |
| Browser submit events / Blocked attempts / Explicit failures | Attempt signals. A blocked attempt is a real submit click or Enter key stopped by native validation. Explicit failures come from the application signal, not an inferred HTTP status. | Event totals may include multiple attempts in one visit. Visits without signals are unknown, not zero. |
The view may also show a trend by day, device breakdown, form/field tables, a ranked Worth a closer look section, and related recordings. Trend dates use UTC. Percentage-point changes compare rates, not visitor volume, and do not establish a cause.
Problems and fixes #
| What you see | What to check |
|---|---|
| Form tracking is missing from the Experience menu | Confirm you selected the intended website, that the dashboard exposes Experience for your account, and that your access role can view it. The feature has a service-side availability gate; ask the account owner or Hitsteps support to confirm entitlement. |
| The report says forms are unavailable or not ready | Check the Form tracking service availability and that Experience is ready for the site. Existing recordings may continue while form reports are unavailable. Try again after the owner/support confirms setup; do not enable unrelated flags or change account settings to work around the gate. |
| The page opens but has no form rows | Confirm the page has a standard supported form, the tracker loads on that path, Experience recording and both form controls are enabled, and there are retained sampled visits in the chosen period. Check consent, browser blockers, path exclusions, date range, and device filter. Data is not backfilled. |
| A form is counted under an unexpected name or appears more than once | Check which page path the form is on and whether the application replaces the form dynamically. Add a stable data-hitsteps-form label or set a safe display name. A form is grouped by page path and identity. |
| A field name is unclear | Use the field's static label or a safe data-hitsteps-field value, or set a display-name override if you have permission. Do not use a personal value as a label. |
| Submission count is lower than the form processor's total | Hitsteps uses a sampled Experience population. Compare the same page and dates, and account for consent, retention, blockers, exclusions, and collection settings. The report is not the system of record for submitted forms. |
| The success rate is a dash or says Signal not configured | The selected visits do not have a usable success signal. Native submits may be estimated only when the browser can attribute the completed navigation. AJAX/single-page submissions may need the optional app outcome integration. A dash is unknown, not 0%. |
| Success appears despite the server rejecting the form | Automatic success detection estimates a completed browser navigation; it does not validate server-side acceptance. Use the optional application outcome signal to report a true result and verify the real submission in your form processor. |
| A field looks problematic but the reason is unclear | Review the visible validation message and the page on a safe test session. Validation, repeat-focus, and exit patterns are evidence for investigation, not user intent or a diagnosis. |
| Counts change sharply between periods | Check whether one range includes today, reaches the retention boundary, triggers the 5,000-visit limit, or has a different sampled/device/consent mix. Compare completed UTC days and similar filters. |
Access, privacy and limits #
Form analytics is separate from form processing. It reports aggregate interaction signals and does not display answers, field values, page text, or request/response bodies. Form analytics does not collect field values. Password fields do not receive inferred captions, and field metadata is limited to interaction signals and safe field identities/labels. Session replay has its own masking controls for password and payment inputs and is a separate collection surface. A form or field display name is still text you provide, so keep it free of personal information.
Experience recording settings also affect the data. Consent, sampling, browser blockers, path exclusions, retention, and collection availability can reduce coverage. The Form tracking view identifies its basis as retained sampled form visits. Its report can be limited to the 5,000 most recent matching visits; narrow the dates or filters when the limit notice appears. Comparisons are withheld when a period is incomplete, today has not finished, or retention limits coverage. No historical sampling rate is available to turn these counts into an estimate of all visitors.
The repository currently applies the website's Experience retention setting to form-report rows; the default in the dashboard is 30 days, but effective customer retention may differ by saved settings and service policy. Confirm the active retention period in the website's current settings and privacy terms before relying on a specific number. This chapter does not provide legal advice. Review privacy, terms, and your own consent setup.
Last checked #
Checked 8 November 2026 using product documentation and source, with public pages and safe sample screens where available. Account-specific access, provider controls and effective limits can differ; use the current message shown for your website. Examples are illustrative.