Customer service and chatbots
Assistants that answer on the website, on WhatsApp and by email from the company's own information, and hand the conversation to a person when they need to.
Artificial Intelligence
Customer service, processes and data: where artificial intelligence takes repetitive work off people's hands, and where it is better left alone.
Artificial intelligence that does work, not demonstrations
Artificial intelligence is only of interest to a company when it takes over work that is done by hand today and that nobody enjoys doing: answering the same question twenty times, copying data from an email into a spreadsheet, digging through a folder for the document that answers a client. That is the starting point for this discipline — the actual task, not the technology.
The four services follow the same rule: you only automate what is already understood. A muddled process, automated, is muddled and fast. So we start by looking at what the team does, we measure where the time goes, and only then decide what goes to the machine, what stays with people, and what is better left as it is.
This discipline nearly always works shoulder to shoulder with development: a support assistant lives inside the website, an automation needs access to the management systems, and the links between platforms are the same ones we have been building for years. What is different is how much decides itself along the way.
When it makes sense to talk to us
Work in this discipline reaches us almost always for one of these reasons:
- the same dozen questions eat the day of someone who should be selling;
- information is being copied by hand from one system to another, every day;
- enquiries come in from several places and there is no way to tell where each one stands;
- the team is already using artificial intelligence tools off its own bat, each in their own way, with no rules and no record;
- somebody promised that artificial intelligence solves everything, and you need to know what is true in your case.
The four services in this discipline
Each service has its own page, with the process, the deliverables and the questions we get asked.
Workflows and processes
Repetitive tasks turned into automatic flows: sorting requests, reading documents, filling in records and alerting the right team.
CRM and integrations
Choosing and setting up the CRM, with enquiries arriving on their own, being sorted, and reaching the right salesperson without anyone copying a thing.
Advice and delivery
Working out what is worth automating, rules of use for the team, and seeing it through — with or without our hands in it.
Tell us which task you want out of the way
You do not need to know which technology to use. Describe the task that keeps repeating, who does it today and how long it takes, and we will tell you whether it is worth automating, what stays out of it, and what it costs. If you would rather talk first, book a meeting or use the contact page.
Common questions about artificial intelligence
The questions people put to us before a job in this discipline starts.
Is this going to replace people on my team?
The aim is to remove repetitive tasks, not jobs. In practice what usually happens is that the same team stops spending the day answering the same thing and copying data, and starts dealing with the work that needs a decision. If the aim is to reduce headcount, say so up front: it changes what gets automated and what has to be made foolproof.
Will the company's data end up in a public model?
Not without your say-so. Each project settles where the data is processed, which supplier is used, and whether the content may be used for training. Some solutions run on European servers, and some tasks we do with no model at all, because it is not worth the risk.
And when the artificial intelligence answers badly?
We start from the assumption that it will. So the assistants answer from the company's own content rather than from what they invent, they say they do not know when there is no answer, and they hand the conversation to a person instead of chancing it. Everything answered is recorded and can be reviewed.
Do I have to change the tools I already use?
Usually not. Most of the work is joining up what already exists. When the current tool allows no connection at all, we say so before starting and weigh up with you whether it is worth swapping or whether it is better to automate a different part of the process.
How long before I see a result?
It depends on the service, but the rule is to start small: one task, one flow, one support channel. That way you measure what changed before widening it, instead of automating the whole company at once and not knowing what worked.
Services in Artificial Intelligence
Customer service and chatbots
Assistants that answer the questions you hear every day, drawn from your company's own content, and hand the conversation to a person when they don't know the answer.
Advice and delivery
An honest assessment of what's actually worth automating, with priorities, ground rules for the team, and delivery to follow, if it makes sense.
CRM and integrations
The right CRM, set up and connected to everything else: contacts that come in on their own, classified, followed up, with nothing copied by hand.
Workflows and processes
Daily repeating tasks turned into automatic flows, with a record of what happened and an alert when something fails.
The only limit is creativity.
Put our team to work, and catch the bug.