Convert abandoned cart
into sales with AI-Voicebot
Challenge
70% of online shopping carts are being abandoned
It’s difficult to engage user afterwards: open rates of emails are 10-20%, SMS is 4-5 times higher but still low
How Voicebot solves it
Calls customer with personalized offer based on their choices
Voice calls have the highest engagement rates
After the call Voicebot sends SMS with details and updates CRM
Integration with any CRM
Support during & after implementation
Test abandoned
cart Voicebot
cart Voicebot
Test our Voicebot: you can go to Apifonica shop, choose some goods and place them in your cart and after that abandon the cart. Once you do that, Voicebot will contact you regarding the goods chosen.
1000+
calls per second
24/7/365
24/7/365
NLP
voicebot understands natural language which boost engagement
4.2 PLN
av. price of 1 lead qualified by Voicebot
150+
languages to talk to your
leads like a local
leads like a local
What is an AI-Voicebot?
This Voicebot uses machine learning and natural language processing to understand
natural speech and respond like a manager would. It automates communication, CRM management, and reporting.
How does it convert abandoned carts?
Voicebot calls a shopper who abandoned a cart, collects feedback and offers a discount or alternative
based on the previous choice. It personalizes and automates communication to boost conversion.
Why is Voicebot the best option?
There are many ways to reach such shoppers. But which is the best? Open rates of emails are the
lowest - 10-20%, SMS is 4-5 times higher, but voice calls have the highest engagement rates.
It takes only 3-5 days to implement such a
Voicebot! Book a free consultation to learn more
Voicebot! Book a free consultation to learn more
Benefits you get with AI-Voicebot
Personalization
AI-Voicebot can help you personalise and automate communication. It integrates
into your CRM and uses webhooks to trigger actions: e.g. user abandons the cart with
certain products and this triggers a call where Voicebot will talk about chosen items.
24/7 availability and speed
Voicebot handles thousands of calls at a time, without lunch breaks and vacations.
Voicebot will need only 10 minutes to process 1000+ customers.
Automation
Voicebot makes a call and updates your CRM automatically. It can also be triggered
by CRM data changes via webhooks: you change pricing and voicebot updates call script
accordingly.
Try it before you buy it
Abandoned
cart call
Voicebot will call each user right after they abandon the cart and still can be converted.
Voicebot can make 1000+ at a time. And each call will be personalized according to the user's
choices. You can listen to an example of such a call.
0:00
0:00
Abandoned cart
free pilot
We offer a free pilot to show you how this all will work for your company and with your
workflows. No commitment or hidden fees. The test lasts 14 days.
63% of marketers see the highest benefit of
personalization on conversion rate increase. Boost
your sales with personalisation via AI-Voicebot!
your sales with personalisation via AI-Voicebot!
FAQ
How much does this solution cost?
We offer 2 different packages and the cost depends on which package you choose
and the number of contacts you want to process. The abandoned cart Voicebot is Voicebot
with Basic Functionality - and it would cost starting from 2800 PLN for 1000 contacts
and integration. There’s also a Professional AI Voicebot with more comprehensive
functionality. You can find details on both variants as well as the exact costs on
our pricing page.
How long does it take to implement Voicebot?
It depends on which variant of AI-Voicebot you choose since we offer 2. First one
with templates - Voicebot with Basic Functionality - can be implemented in just a few days.
You get a fully functioning AI-Voicebot with editable script templates, speech synthesis and
recognition, webhooks, SMS triggered by action/answer, local phone number and reporting.
In certain cases customers need something very customized and we provide that as well. There’s a custom pro version with conversational scenarios featuring multiple sequences of events/replies. Such Voicebot can consult in multiple areas, understand multiple intents and its implementation may take from 3 to 5 weeks.
In certain cases customers need something very customized and we provide that as well. There’s a custom pro version with conversational scenarios featuring multiple sequences of events/replies. Such Voicebot can consult in multiple areas, understand multiple intents and its implementation may take from 3 to 5 weeks.
Does Apifonica provide implementation support?
We provide full support from initial idea to the release of the project: we’ll set it all up
for you. Moreso, our expert team provides initial analysis of your business workflows to ensure
you get the optimum solution based on your problems, needs and goals. Depending on the complexity
of your business process we provide adaptation and integration of our solution into your business
processes and ensure smooth integration with your CRM system.
How does vb know how to react? And what to say?
So, let’s imagine a situation, where a bank customer calls a support, complaining about their
credit card, which is not working. There are different ways to talk about this issue:
Voicebot can understand all these different variants, including interjections through intents. An intent is basically what the user wants or their intention. In this case the intent is the - ‘blocked card’. And all the different ways, in which the user can express it, are called the ‘training phrases’. In other words, they are the predicted phrases that the user may say. When the user’s utterance overlaps with, or matches the intent that the voicebot knows, then the voicebot knows how to react and what to say.
- my card is not working;
- my card is blocked;
- I couldn’t make a payment;
- something is wrong with my card;
- why can’t I make a payment
Voicebot can understand all these different variants, including interjections through intents. An intent is basically what the user wants or their intention. In this case the intent is the - ‘blocked card’. And all the different ways, in which the user can express it, are called the ‘training phrases’. In other words, they are the predicted phrases that the user may say. When the user’s utterance overlaps with, or matches the intent that the voicebot knows, then the voicebot knows how to react and what to say.
How AI helps to understand the end user better and what machine learning has to do with it?
It all comes down to the process of intent matching, where AI based machine learning is
applied. Machine learning mechanism compares the intent of the user with the intents that the
voicebot knows and finds the best match. How does the matching process work?
Machine learning algorithms calculate the confidence score for each intent that the voicebot knows. The confidence score is marked on a scale from 0 to 1. Where 0 means that the match is completely uncertain, and 1 means that the match is completely certain. The vb needs to find the user’s intents amongst all the variants. For this the threshold is defined in the process of the voicebot creation and testing. The threshold does not define how good the vb is in finding the variant. But it helps us to set how strict we are in letting the vb find the right intent. Opinions vary on what the threshold should be like, usually it’s from 0.3 to 0.7 depending on the complexity of Voicebot. So, again, In other words, machine learning classifies which intent that the voicebot knows is the most similar to the intent said by the user and this is how Voicebot understands people.
Machine learning algorithms calculate the confidence score for each intent that the voicebot knows. The confidence score is marked on a scale from 0 to 1. Where 0 means that the match is completely uncertain, and 1 means that the match is completely certain. The vb needs to find the user’s intents amongst all the variants. For this the threshold is defined in the process of the voicebot creation and testing. The threshold does not define how good the vb is in finding the variant. But it helps us to set how strict we are in letting the vb find the right intent. Opinions vary on what the threshold should be like, usually it’s from 0.3 to 0.7 depending on the complexity of Voicebot. So, again, In other words, machine learning classifies which intent that the voicebot knows is the most similar to the intent said by the user and this is how Voicebot understands people.