Тёмный

Using ChatGPT to Automate Cold Calling for Sales Leads [with Bland AI] 

Подписаться
Просмотров 6 тыс.
% 235

Get 25% off Bland AI:
app.bland.ai/signup?ref=763d7133-0070-45e9-92ac-bf7c83080a62
Automate cold calling with ChatGPT and Bland AI. I'll walk you through how to setup and use the program I wrote to run an automated cold calling campaign that forwards warm leads directly to your phone.
Files & Links from Video:
BlandAI: app.bland.ai/signup?ref=763d7133-0070-45e9-92ac-bf7c83080a62
Colab for Python Code: colab.research.google.com/drive/14CoMFyWhDYz75lxguOaXzckyt7Vs2jsP?usp=sharing
Other Videos:
AI Phone Scheduler - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-bwhBlqhbUrY.html
GPT that makes Calls - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zm6SdRW5TGU.html
Phone Negotiating Bot - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-k0hXggYd7do.html
Fine-Tune Llama-2 - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-u0hpZ3HRs1c.html
Fine-Tune ChatGPT 3.5 - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-8Ieu2v0v4oc.html
Build an AI Texting Bot - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-dr7q0cDujjc.html
Fine-Tune ChatGPT 3 - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-eS7o9IfkJxI.html
Embed Data with ChatGPT - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-EQ9Galv242c.html
Build a chatbot - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zVgJiQT9DHU.html
Chapters:
0:00 - Intro
2:00 - Setup Excel Data
3:01 - BlandAI
3:54 - Setup Python Code
6:00 - Explain Call Payload
12:24 - Call Status
13:00 - Generate Call Report
14:37 - Emergency Stop
14:50 - Running it
Music:
"jiglr - Odyssey" is under a Creative Commons (CC BY-SA 3.0) license
@jiglr
Promoted by BreakingCopyright: breakingcopyright.com
"Loopster" Kevin MacLeod (incompetech.com)
Licensed under Creative Commons: By Attribution 4.0 License
creativecommons.org/licenses/by/4.0/
Music track: Space by Damtaro
Source: freetouse.com/music
Royalty Free Music for Video (Safe)

Наука

Опубликовано:

 

17 мар 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 61   
@ernestosazo299
@ernestosazo299 5 месяцев назад
Wow! Thank you so very much. Truly outstanding very grateful for the detailed explanation and description of python code and the possibilities to change it. Exactly what I have been looking for, subscribed and truly looking forward to see everything you put out. Thank you, looking forward to showing my appreciation.
@tech-at-work
@tech-at-work 5 месяцев назад
Thank you for the awesome comment! Glad it was useful!!
@aimattant
@aimattant 4 месяца назад
Where's the demo call?
@5-MinuteBedtimeStories_
@5-MinuteBedtimeStories_ 5 месяцев назад
Legend. Need more videos ff your AI voice agents utilities!
@baeshayoung2544
@baeshayoung2544 7 месяцев назад
Such a good tutorial! Thanks for sharing 🙏
@tech-at-work
@tech-at-work 7 месяцев назад
Thanks! Glad it was useful!
@ChessDon
@ChessDon 6 месяцев назад
Awesome video. Just subscribed. Would love to see a video on you setting up a inbound voice agent with bland!
@tech-at-work
@tech-at-work 6 месяцев назад
Awesome! Thank you, yea I’ve had that request a few times, I’ll look at making a video for it
@ChessDon
@ChessDon 6 месяцев назад
@@tech-at-work a unique spin that you could put on the video is showing how to use an inbound agent to book a call AND leveraging bland AIs dynamic data objects in the build. For example maybe the inbound agent books an appointment but first checks that there is availability in the calendar. I've seen other videos on setting up appointment bookings but I've yet to see one where the agent first checks that there is availability in the calendar at that time
@tech-at-work
@tech-at-work 6 месяцев назад
You’re reading my mind :)
@ChessDon
@ChessDon 6 месяцев назад
@@tech-at-work i can't wait! My post notifications are turned on! Your channel is going to blow up someday soon I'm sure. This content is top notch!
@tech-at-work
@tech-at-work 6 месяцев назад
Thanks for the support!! I really appreciate that
@ritaverissimo1720
@ritaverissimo1720 27 дней назад
Hi, awesome video, thanks ! 2 questions: - Is it possible to make calls in French and Spanish aside English? - Is it possible to make an automation? Like each time there's a new contact added to the csv it gets updated and the person gets the call. TIA 😊
@AnewViewTeam
@AnewViewTeam 2 месяца назад
Great video, thank you!!! Question for you - how do you customize the AI with additional sentences & responses? For example, you wrote “first sentence”. How would I add responses and a 2nd or 3rd sentence?
@rpsingh18
@rpsingh18 5 месяцев назад
This is very helpful! It worked.
@amiradj9695
@amiradj9695 7 месяцев назад
Another bomb! Thanks for sharing.
@tech-at-work
@tech-at-work 7 месяцев назад
Thanks for the support!
@NevadaHomeBuys
@NevadaHomeBuys 5 месяцев назад
Wow, this is great! I am a Real Estate agent looking for a bot to do lead calling from people who have requested more information. I would have liked to hear the final product of your AI Bot making that first call. Curious to see how it flowed. Do you offer assistance to create the bot??
@wendelllaidley1065
@wendelllaidley1065 6 месяцев назад
@tech-at-work Do you have suggestions for 3rd parties I could hire to help implement what you have put together in this video? I am a business owner lacking developer skills and would like to contract with someone to stand up this functionality (linked to High Level CRM). Any suggestions? Many thanks!
@truewarriorfitness
@truewarriorfitness 4 месяца назад
Is there a 4o upate video coming soon?
@tech-at-work
@tech-at-work 4 месяца назад
I will work on this! Thanks for the suggestion
@Kevinsmithns
@Kevinsmithns 5 месяцев назад
Is there an update to this where I can train the ai go be more specific with its training prompt of what to say? And do you have a way for incoming calls also with pinecone so it remembers customers when they call back
@fastbreak9680
@fastbreak9680 7 месяцев назад
brother, in order to call, should I top up my blanc ai? And bro, where can I see the transcript of the entire conversation? If I'm not mistaken, then it should be in the table, right? with the answer to who answered the call, was the customer interested and so on
@tech-at-work
@tech-at-work 7 месяцев назад
If you’re going to run the full batch, you’ll need BlandAI credits, yea. You can have it provide the transcript for each call in the questions, yes. I would set it up like “give me the exact transcript of what was said on the call” and make sure you set it as a string for the response. I would also make sure and mention you want a transcript in the goal portion.
@christianstill.6654
@christianstill.6654 7 месяцев назад
This is TOO GOOD! 😂😂
@tech-at-work
@tech-at-work 7 месяцев назад
Thanks for the compliment! Every Like helps me make more videos :)
@christianstill.6654
@christianstill.6654 7 месяцев назад
@@tech-at-work Lol I will like in a sec. Quick Question: can you make a full tutorial on how to link call data (like appointments) to google calendars, get call recordings, etc? A lot of people aren't programmers, so going through the documentation on BlandAi could be confusing! Thanks in advance!
@christianstill.6654
@christianstill.6654 7 месяцев назад
@@tech-at-work Also how do we get answers to more questions throughout the call? that would be helpful too
@tech-at-work
@tech-at-work 7 месяцев назад
Thanks for reconsidering! And yeah I’ll try to put up a video on linking appointments to Google calendar, I’ve received similar requests before
@tech-at-work
@tech-at-work 7 месяцев назад
You can add more than 4x questions for the report it creates from the calls, just keep adding lines below the current questions (inside the square brackets) in the report section!
@IAppsSAS
@IAppsSAS 7 месяцев назад
Hey there. I just got this issue: Failed to send batch {"status":"error","message":"Errored on very first call and stopping execution so that it can be fixed. Error messages: Request failed with status code 400"} Do you've any idea how could I solve it? thank you!
@tech-at-work
@tech-at-work 7 месяцев назад
Make sure the first phone number (in cell B2) is real, otherwise it’ll error out
@fastbreak9680
@fastbreak9680 7 месяцев назад
same problem bro. do you find solution??
@tech-at-work
@tech-at-work 7 месяцев назад
It appears to only work for phone numbers in countries supported by BlandAI at the moment. They have an international-rollout section of their Discord, they update there when new countries and languages are supported
@IAppsSAS
@IAppsSAS 7 месяцев назад
Hey there, I just got this issue: Failed to send batch {"status":"error","message":"Errored on very first call and stopping execution so that it can be fixed. Error messages: Request failed with status code 400"} Do you have any idea about how could I solve it? Thank u sir
@tech-at-work
@tech-at-work 7 месяцев назад
I had that issue if the phone number i put in wasn’t real, try updating the first phone number with your phone number in cell B2
@IAppsSAS
@IAppsSAS 7 месяцев назад
@@tech-at-work Indeed it was, thank u so much. However the problem is that it only allows up to 11 digits in a phone number, and for Colombia, my country, phone numbers have 12 digits. Actually you can see a support ticket talking about that on Discord. Hopefully they will fix it soon!
@tech-at-work
@tech-at-work 7 месяцев назад
Oh! Yeah I should have specified this was setup for the US, I’m not as knowledgeable on the international use :/
@DylanHaines-t2r
@DylanHaines-t2r 5 месяцев назад
A few things I am having trouble with: The voice ID doesn't change even though I change the ID from a voice ID straight from Bland. Anything that could be causing this? It also keeps calling the prospect "Name" even though I have an entered name in the spreadsheet
@DylanHaines-t2r
@DylanHaines-t2r 5 месяцев назад
I think it is because I am trying to add the "name" to the AI's first line of speech. I have it as "Hey, is this {{name}} ?" and it keeps just saying "name" rather than what is on the spreadsheet
@tech-at-work
@tech-at-work 5 месяцев назад
Make sure you have the name variable defined in the call_data section and your column titled correctly. Once it’s defined you have it formatted correctly in the first sentence
@tech-at-work
@tech-at-work 5 месяцев назад
Bland ai has changed the way they handle voice id now, you should be able to copy the id from their dev portal or they have just names you can list like maya, mason, ryan, etc and they use just “voice” now
@DylanHaines-t2r
@DylanHaines-t2r 5 месяцев назад
@@tech-at-work I have it on there just as you have it "name": row {'Name'}, and the first sentence I have is "Hey, is this {{name}} ?", but it still just keeps saying "name" rather than what is in the spreadsheet
@DylanHaines-t2r
@DylanHaines-t2r 5 месяцев назад
Ok so when I ask it, "who are you looking for?" it says the correct name. so it knows who it is calling, it just wont say it on the first sentence for some reason
@funnyprankforlaughter1388
@funnyprankforlaughter1388 7 месяцев назад
You are doing awesome sir ..😊
@tech-at-work
@tech-at-work 7 месяцев назад
Thanks!!
@fastbreak9680
@fastbreak9680 7 месяцев назад
Bro can i use that if im from Kazakhstan 9
@tech-at-work
@tech-at-work 7 месяцев назад
I’m not sure it will work in Kazakhstan, I only tested it in the US
@wanderingpanchhi
@wanderingpanchhi 6 месяцев назад
Great tutorial. The BG music is annoying. Drop that?
@johnmarino7897
@johnmarino7897 4 месяца назад
Or you can just upload your CSV leads to bland and it will call them.
@curiouscoding3130
@curiouscoding3130 5 месяцев назад
it does not work it doesn't call from sheet ,
@tech-at-work
@tech-at-work 5 месяцев назад
Where are you getting an error?
@fastbreak9680
@fastbreak9680 7 месяцев назад
Bro i encounter with that issue KeyError Traceback (most recent call last) /usr/local/lib/python3.10/dist-packages/pandas/core/indexes/base.py in get_loc(self, key, method, tolerance) 3801 try: -> 3802 return self._engine.get_loc(casted_key) 3803 except KeyError as err:
@tech-at-work
@tech-at-work 7 месяцев назад
its trying to access a column in your data that doesnt exist. Make sure you have the column titles correct and referenced in the code the same.