XYZ School is a well-known school for game developers. Its target audience consists of young and ambitious people who are interested in or already work in IT and graphic design.
To attract customers, the school used the tool of loot boxes (boxes with prizes) at the registration stage — they contained promotional codes for discounts of various sizes. In order to win a larger gift, many registered with different e-mail addresses several times.
From XYZ, we received the task to segment the database and offer to use a south africa phone number data promotional code for the company's main courses. When the decision was made to delegate the call — there were 5 days left until the end of the campaign. The bot had to filter leads from a base of 9,000 to give managers only warm interested clients.
School for game developers XYZ School
How did they work and what did they face when calling
Script
First of all, we analyzed the company's target audience and developed 2 variants of the dialogue structure. We adapted to the audience: these are people under 30 who do not like long phone conversations, want to know who is calling and why, and understand the value of the offer well. Therefore, we tried to make the dialogue as simple and fast as possible, and the announcer's voice youthful, lively and energetic.
The main goal of the script was to encourage people to take advantage of the discount and buy one of the offered courses.
Testing
At the stage of A/B tests, we found out that customers distrust our calls. Company XYZ positions itself as "the first company where there is no telephone sales". Because of that, people asked again and wanted to make sure that these are not fraudsters.
In order to fix this and not lose the audience's trust in the school, we decided to urgently rewrite the script. Usually we don't advertise bots, because people are biased towards bots for calls. But for this target audience, everything worked the other way around, and we left a version of the call in which the bot immediately introduced itself to clients as the virtual assistant XYZ.
Calls to clients
We called applications in several stages over 4 days. This was done in order to evenly distribute the burden on managers, because warm leads should have been processed immediately.
A report with the results of the call was sent to the customer daily, and hot and potential applications were immediately separated. All clients were segmented by categories:
Manager's call
Relevant, but later
Training is already underway
Already bought
Not interesting
Delete number
So that the manager does not lose the thread of the conversation, in each section there were verbatim answers of clients recorded in the conversation with the bot.
The results of a 4-day bot call to sell a course