Most articles about business automation are written by people who already understand the topic, for people who already understand the topic. They get straight into Python, APIs, webhooks, and "low-code platforms". If you are not technical, the article is useless about three paragraphs in.
This is the version that does not assume any of that. If you run a UK small business, you have heard the word "automation" and you suspect it might apply to your operations, this is the explanation.
No jargon. No selling. Just what it is, what it costs, and how to decide whether you need it.
What "business automation" actually means
Imagine the work somebody on your team does every Monday morning. They open three pieces of software, copy a number from one into another, format a spreadsheet, save it somewhere shared, and email a report. Same task, every Monday, for two years.
Business automation is replacing that work with a piece of software that does the copying, the formatting, the saving, and the emailing automatically. Not a person doing the same work faster. Not a SaaS subscription that promises to "streamline your operations". A piece of code, written specifically for your workflow, that does exactly the same task without anyone touching it.
That is the whole concept. The complications are about what the code actually does, what it connects to, and how reliable it is. The core idea stays the same: a person was doing repetitive work; now a script is.
What "the bureau builds it" actually looks like
A typical custom automation project from Orchestrix has four steps, in plain English:
Step one: a conversation. Forty-five minutes, mostly spent listening to whoever currently does the work. The goal is to understand exactly what happens today, including the bits that are not written down anywhere and only exist in someone's head.
Step two: a written spec. The bureau writes down what is going to be built, in plain English, so you can read it and confirm it is right before any code is written. This is the bit most agencies skip; it is the bit that prevents the wrong thing being built.
Step three: building. The bureau writes the actual code. Usually Python, sometimes Node.js. The code lives on a server (yours or one the bureau hosts) and runs on a schedule, when triggered by an event, or both. You do not have to look at the code; you just have to confirm that the output is what you wanted.
Step four: handover. The bureau delivers the build, with documentation written in plain English. You get a copy of the code, the documentation, and a 30-minute call where it is explained. After that, the build runs whether anyone is watching or not.
The total timeline is usually two to six weeks. The total cost is usually £2,000 to £8,000 for a focused build, fixed-fee, agreed before the work starts.
What it can replace, in plain examples
Some real workflows the bureau has automated for UK SMBs:
- Every time a new sale closes in the CRM, an invoice gets generated and sent to the customer, the customer gets added to the email newsletter list, the project gets created in the project tool, and the team Slack channel gets a notification. Used to be 15 minutes per sale, manually. Now zero.
- Every Friday at 6pm, last week's sales numbers, this week's pipeline, and the running total against monthly target get pulled from three systems, formatted into a dashboard, and emailed to the leadership team. Used to be three hours every Monday morning. Now zero.
- Every time a customer fills in the contact form on the website, the message gets routed to the right person based on what they asked about, a calendar invite gets created if they want a call, and a record gets logged in the CRM with the conversation tagged correctly. Used to be a junior team member triaging email all day. Now they do not.
- Every quarter, the compliance evidence (DBS records, training certificates, insurance documents) gets pulled together from four sources into a single PDF for the auditor. Used to be a full day of someone's work. Now it generates overnight.
These are not exotic. Almost every UK SMB has at least one of them sitting around as manual work somebody does every week without thinking about it.
What it cannot replace
The bureau is honest about the boundary:
- It cannot replace decisions that need human judgement. The phone call where you negotiate scope. The difficult conversation with a customer who is unhappy. The decision about which job to prioritise this week.
- It cannot replace relationship-led work. Account management, sales, customer success.
- It cannot replace creative or design work. Writing the marketing copy, designing the brand, drafting a proposal.
- It cannot fix a broken process. If the underlying workflow does not make sense, automating it just makes the broken thing happen faster.
The right way to think about it: automation is for the work where the 100th instance is the same as the first. Anything that needs daily judgement, taste, or human relationship-building is not the right shape.
What it costs (the maths in plain numbers)
Three rough categories:
Small automation: £1,000 to £4,000 fixed-fee. A focused script that connects two tools and replaces one specific manual task. Usually ships in two to three weeks. Pays for itself within months if it replaces an hour or more per week of someone's time.
Medium automation: £4,000 to £10,000 fixed-fee. Multiple integrations, conditional logic, scheduled jobs, error handling, monitoring. Usually ships in three to six weeks. Replaces the kind of workflow that currently eats half a day of someone's time per week.
Larger build: £10,000 to £25,000+ fixed-fee. A custom internal tool, a customer portal, a dashboard system. Ships in six to twelve weeks. Replaces a workflow that currently requires a dedicated person or causes ongoing operational friction.
For a UK SMB paying a £35,000 salary to someone whose job is partly admin work, even a £10,000 build that frees up half their time pays for itself in eighteen months and continues to save money every year afterwards.
How to know if you need it
Three signals that suggest you might:
- Someone on your team is regularly working evenings or weekends to keep up with admin work that is not their actual job.
- You have a load-bearing spreadsheet that nobody fully understands but everyone depends on.
- You can name a specific task that takes hours of human time every week and that follows the same pattern every time.
If two or more of those are true, the 15-minute triage is the cheapest way to find out whether automation is genuinely the right answer. The triage is free, takes 15 minutes, and ends with an honest fit assessment. Roughly half the time the answer is "no, this is a hire problem" or "no, use Zapier" or "no, your process needs sorting first". That is the point of the conversation.
If the answer is yes, the operational audit is the next step. It is a paid, fixed-fee engagement that maps your operations and produces a written punch list with fixed-fee proposals for each fix. The audit fee credits in full against any build that follows, so if you do build with the bureau, the audit costs nothing in net terms.
What to expect at the triage
The conversation is genuinely a conversation. There is no slide deck. There is no demo. There is no follow-up sales sequence. You describe the workflow that is bleeding the most time. The bureau asks questions. By the end of fifteen minutes, you know whether the workflow is a candidate for automation, what it would cost to fix, and what the alternatives are.
If you have read this far and the workflows in your operation came to mind, that is usually a good sign that the conversation would be worth having.