Assistants, automation and models over your own data, aimed at the manual steps that are costing you the most, with a person checking the answer.
What we measure
15 hrsoffice time saved a week · removals firm
Someone on your team spends their day answering the same questions, reading the same documents, and typing the same things into the same systems.
One client, ten weeks
A removals company whose office team read every quote request by hand while customers waited a day or more. We built an assistant that reads each request, asks for anything missing, and drafts the reply. Nothing reaches a customer until somebody in the office has checked it.
Time to send a quote
BeforeAbout 26 hours
After3 hours
What you can answer afterwards
Every answer checked against a test set before it reaches a customer.
You get
An assistant that reads your policies, catalogue or records and answers customers or staff with a citation to the source.
You get
The reading, sorting, drafting and re-typing that fills a working day, handled before anyone arrives.
You get
A feature your customers use, built into the website or app you already have.
You get
Every answer checked against a test set before launch, and measured after it, so you know the error rate rather than hoping.
In more detail
The useful ones answer from what you actually know, admit when they do not, and pass the question to a person before the customer gets angry.
Somebody in your business spends Thursday moving information from one screen to another. That is the work worth automating first.
What we actually set up
The same work in technical terms, so you or your developer can check it against anyone else’s quote.
FAQ
Left alone, any model can. That is why every answer comes with its source, why anything customer-facing is checked against a test set before launch, and why a person approves anything that goes out under your name until the error rate is one you have seen and accepted.
Into accounts you own, under terms that do not allow it to be used for training. We will show you the terms before anything is connected, and we do not connect anything you have not seen.
No. Most useful work is an assistant reading a few hundred pages you already have, or automating a step someone does forty times a day. Large datasets are the exception, not the requirement.
Usually pennies per task, and we measure it. You will know the cost of every answer before you decide whether to keep it.
The documents or systems the work touches, someone who knows how the job is done today, and a few hours to build the test set with us. The test set is the most valuable thing in the project, and it has to come from you.
Your turn
Tell us about your business in a few minutes. We will reply within a working day to say whether it is measurable, roughly what it would take, and whether we are the right people for it.
