Bot Detection
Bot Detection is currently available to select customers. Reach out to [email protected] if you'd like access.
When sharing a Long-Form Survey link via email, security software on the end-user’s device may click through the link and complete the survey. This would result in survey responses for the Long-Form Survey being a mixture of real end-user responses and bot responses. To filter out bot responses from human responses, you can enable an invisible reCAPTCHA (powered by Google) that is added to Long-Form Survey links. The reCAPTCHA examines the behavior of the survey taker as they complete the survey and scores each survey taker on whether they are likely a human or a bot. The reCAPTCHA score is saved with the user’s survey responses in Sprig and can be used to filter out suspected bot responses, ensuring the integrity of the data collected.
How to Enable
Bot Detection is off by default. To enable it:
- Go to Settings > Team > Bot Detection.
- Click the toggle to enable Bot Detection.
Once Bot Detection is enabled, the reCAPTCHA analysis will run as respondents complete a Long-Form Survey.
If Bot Detection is enabled, a disclaimer that says: “This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.” will appear at the bottom of Long-Form Surveys and be visible to respondents. This is required by Google and cannot be removed.
When disabling Bot Detection, it may will take a few minutes for the disclaimer to disappear from the respondent view of the Long-Form Survey.
Filtering Suspected Bot Responses
When Bot Detection is enabled, there will be a new toggle on the Summary page and on the Responses page of a survey that allows you to exclude suspected bot responses from the results.
On the Responses page, there will be a note on individual responses that says “Suspected bot” to warn that the response likely came from a bot.
When exporting study data via CSV, suspected bot responses will be included, but you will have the option to include reCAPTCHA scores in the export. The score will be in a new column labeled “captchaScore.” Scores range from 0.0 to 1.0. The higher the score, the more likely that the response is from a human. Learn more by reviewing Google's interpreting reCAPTCHA scores page.
Suspected bot responses are sent to Slack, webhooks and the API.
Updated about 10 hours ago