Automated Spam and Malicious-intent filtering for Conversations, "Mark as Spam" feature for spam messages
in progress
P
Perinbhan Ponnampalam
When we get spam messages in our inbox. Can we have in the dropdown shown a note that this is spam and that will also remove the contact from our CRM or block them? Some clients get a lot of spam messages from social media.
Photo Viewer
View photos in a modal
Log In
Jehan Zouak
Merged in a post:
Let Web Chat AI mark solicitations as spam so they don't trigger New Lead notifications or create contacts
H
Hoop Jumper
We manage Web Chat with the AI Chat Receptionist for real estate clients, and we're getting a steady stream of spam and solicitation chats (SEO offers, guest posts, web design offers). We configured the AI to recognize these, send a closing message, and stop replying, and that works well when the spammer gives no contact details.
The problem is when a spammer types their own email address or phone number into the chat. Vendasta then automatically creates a CRM contact and sends the "New Lead from Web Chat" notification to assigned users and followers. Support confirmed (ticket #1a0c54df9be4cf2f) that native lead capture is hardcoded to parse any email or phone in the chat, and that the notification cannot currently be suppressed by tags, capability rules, or automation conditions.
Our team then wastes time on fake lead alerts, and clients lose trust in the notifications, which makes real leads easier to miss. The AI summary that explains it's a solicitation only appears about five minutes after the contact and notification are created, so the alert has already gone out.
What we'd like to see
A setting to hold or suppress the New Lead notification when the AI identifies a chat as spam or a solicitation.
A way for the AI to mark a conversation as spam or disqualified, through a built in capability or tool, so it doesn't create a contact or notification.
A Spam or Not a Lead status or tag that can be applied at the widget level, and that automations can use as a condition.
Access to the AI summary text as a variable in automation conditions, so partners can filter on it without relying on an AI categorization step after the fact.
Why this matters
Partners running Web Chat for lead generation clients need real leads to be trusted and spam to stay out of the CRM. Manual cleanup and unfollowing threads don't scale across multiple client accounts. Even one of the options above would remove most of the noise.
Thank you for considering this.
M
Mustafa Sharif
Any updates on this feature?
Rylan Morris - Vendasta
Merged in a post:
Close (or take other actions on) an inbox message after a period of inactivity
B
Brian McGuire
We get a lot of spam calls and these really fill up the inbox. Each one has to be closed individually. I would like to create an automation that will mark a conversation as closed after a period of inactivity.
N
Nikki Buenaventura
When customers receive spam messages, can we have a delete and report as spam option in the dropdown filter or in the kebob? We have customers reporting receiving multiple spam messages from social media.
Photo Viewer
View photos in a modal
Vendasta Team
An update: We are still working on an AI analysis engine for conversations, to identify spam / malicious / off-topic conversations and allow for different actions based on detection, like stopping responding, auto-filtering, etc. Thanks for patience as we get this ready for wider use, while ensuring no lead is left behind.
Vendasta Team
Merged in a post:
Customize behavior of AI Chat Receptionist when engaged by users with malicious intent
J
Jim
In addition to being able to filter spam, it's also important for the AI Chat Receptionist to know how to respond when a user tries to engage it in inappropriate conversations. For example, when a user submits a riddle to lead the AI employee to determine their name is not actual contact data but inappropriate language, the employee should recognize that, flag it, and not log the conversation and contact as a lead, but it should also know how to shut down the conversation rather than allowing the user to continue engage them.
T
Tom Lombardi
This is very important. We get trolled everyday
P
Perinbhan Ponnampalam
Hello Team,
Also please include delete options for any spam incoming messages on the Conversation Tab
S
Scott Rutherford
Will this include spam from all methods (e.g., chat, voice)? If not, please consider. Thanks in advance.
Vendasta Team
updated the status to
in progress
We're internally testing automated AI-powered spam detection. When ready, we plan to add user controls around this too, in case messages were erroneously detected or not detected, user can fix afterwards.
Load More
→