Customer service
How to answer faster and better without growing your team.
Telegram vs WhatsApp for customer service: why you end up with both
Almost every comparison asks you to choose between Telegram and WhatsApp. In a real business the question arrives late, because both channels are already open and each one is watched by someone different. What actually changes between them, and what to do with both.
Read more →Handoff rules: which topics your AI should never answer
Most people configure what the AI has to answer and skip the opposite, which pays off more. The five kinds of question worth always escalating, how to build the list with your team, and why escalating too much at first is right.
Read more →Will AI replace customer service jobs?
This is the question that comes up before you show any of this to your team, and it deserves an answer with data rather than a reassuring phrase. What the global studies project, what happens in a twenty-person business, and which tasks change hands.
Read more →Answering customer messages after hours without your team working nights
The message lands at eleven at night. Your team finished five hours ago. If nobody answers, the lead goes cold; if somebody answers, you are asking for hours you never agreed. The three possible ways out and what each one costs.
Read more →Your customers' data in the hands of an AI: seven questions for the provider
When you connect an AI to your WhatsApp, your customers' conversations pass through a third party. In Argentina that is governed by Law 25.326, and the data protection authority has published recommendations on the responsible use of artificial intelligence. What it means in practice and what to ask a provider before signing.
Read more →Handling patient messages on WhatsApp without breaking privacy law
A lab or a medical centre that answers on WhatsApp is handling sensitive data, even if nobody calls it that. In Argentina that sits under Law 25.326 on personal data and Law 26.529 on patient rights; wherever you are, the law has a different name and the criterion is the same. Which queries you can automate without touching anything clinical, and which have to reach a person.
Read more →WhatsApp for building management: 312 conversations a month without losing one
A building management company that handles residents over WhatsApp moves more volume than it looks: 312 conversations and 4,137 messages a month in a real case we measured. Half of those issues resolve without anybody on the team stepping in. 40% need a person, and knowing which is which is the whole job.
Read more →Transcribing WhatsApp voice notes to text: why a phone is not enough for a business
Seven billion voice notes are sent on WhatsApp every day. When a business handles customers there, somebody has to listen to all of them, in order, to find out what each customer asked for. WhatsApp already transcribes, but it does it on your phone and one message at a time. What changes when the transcription lives in the business's conversation.
Read more →87% of customers demand a way through to a human
The rejection you are measuring is not aimed at AI. It is aimed at AI with no emergency exit. The two Gartner numbers that look contradictory and say the same thing.
Read more →Do you have to tell customers they are talking to an AI?
The legal answer and the practical answer rarely line up. In Argentina, where we looked at it closely, no rule in force requires the disclosure. There is still a good reason to make it.
Read more →Customer service for online stores: who answers the question that comes before the purchase
US retail e-commerce reached $326.7 billion in the first quarter of 2026 alone, 16.9% of everything sold at retail, growing at 9.8% against 3.9% for retail overall. Behind almost every one of those orders there was an earlier question: size, shipping, stock, warranty. That enquiry decides the sale, and it is the one that usually goes unanswered.
Read more →Who handled that customer? The case nobody owns is the one you lose
When three people share the same channel, a problem shows up that has nothing to do with speed or tone: nobody knows who owns each case. The result is not a weak reply, it is no reply. How assignment and a thread you can read end to end fix it.
Read more →