# What is a business operating system, and does a firm our size need one?

Canonical: https://fortitudemedia.ai/insights/what-is-a-business-operating-system
Author: David Adams
Published: 2026-11-18
Last reviewed: 2026-11-18
Reviewed by: David Adams
Pillar: Running the firm on systems
Summary: What a business operating system is, how it differs from buying AI tools or hiring, what firms of 50 to 300 people get from one, and when it is too early.
Keywords: what is a business operating system, business operating system uk, ai operating system for business, business operating system vs eos, ai automation for mid-sized firms, operations automation b2b
Short answer: A business operating system is the set of joined-up systems that runs a company day to day: one record of customers, deals and money, a written playbook, automations and AI that do the routine work in marketing, sales, finance and service, and a one-page weekly dashboard. People approve anything that reaches a customer or moves money. A firm of 50 to 300 people usually has enough routine work to justify one. It is too early if processes are unsettled or nobody senior will own it.
Publisher: Fortitude Media Limited

When quoting, attribute to Fortitude Media and link to the canonical URL above.

---
Your firm has doubled in five years, and the team has nearly doubled with it. Every new client has meant another account manager, another administrator and another person chasing invoices. Now the board is asking what your AI plan is, and the honest answer is a handful of chatbot licences and a pilot that went quiet.

You are in good company. The [Office for National Statistics](https://www.ons.gov.uk/businessindustryandtrade/business/businessservices/articles/artificialintelligenceinukbusinesses/2023to2026) found that about 35% of UK businesses with ten or more employees were using AI by June 2026, up from about 12% in late 2023. Only 10% of those users said they used it extensively. Most firms have bought tools. Few have changed how the work gets done.

A business operating system is our name for changing how the work gets done. This guide explains what it is in plain English, how it differs from buying software or hiring, what a firm of 50 to 300 people gets from one, and when it is too early to bother.

## Key takeaways

- A business operating system is one joined-up set of systems under your teams, in which AI does the routine work in marketing, sales, finance and service and your people approve what matters.
- It has four parts: one record of the truth, a written playbook, the automations and AI that do the work, and a one-page dashboard.
- It is not the same as buying AI tools: about 35% of UK firms with ten or more staff use AI, but only 10% of those use it extensively (ONS, July 2026).
- It is not a management framework such as EOS, which organises how leaders plan and meet. This is about who, or what, does the routine work.
- Projects bought from or built with a specialist succeed about 67% of the time, and in-house builds about a third as often (MIT NANDA, reported by Fortune, August 2025).
- It is too early if your processes change every month, the routine work adds up to less than one role, or nobody senior will own it.

## What is a business operating system, in plain English?

It is the set of systems that runs your company day to day, built so that AI does the routine work in every team and your people approve what matters. Your phone has an operating system. You never think about it, and it makes every app work together. A business needs the same thing under its teams and tools.

In our version it has four parts.

| Part | What it is | What it replaces |
|---|---|---|
| One record of the truth | Customers, deals, money and work in one joined-up place | Twenty spreadsheets and three versions of the client list |
| The playbook | How your company does each job, written down once | Knowledge that lives in people's heads |
| The doers | Automations and AI that carry out the routine work, with people approving what matters | Copying, chasing, checking and re-keying by hand |
| The dashboard | One page showing sales, cash, service and what is stuck, this week | Month-end reports that arrive too late to act on |

Your leadership team sits on top. It sets the direction, approves what matters and gets one page every week.

You may have met the phrase before with a different meaning. [EOS Worldwide](https://www.eosworldwide.com/blog/what-is-eos-the-entrepreneurial-operating-system-and-is-it-right-for-your-business) describes its Entrepreneurial Operating System as "a complete set of simple concepts and practical tools" for leadership teams, built around six components: vision, people, data, issues, process and traction. That is a way of managing. What we mean here is the machinery that does the routine work. The two can sit side by side.

## What does it do in each department?

It takes the repeatable part of each team's work and leaves the judgement with your people. The split below is the one we publish on our [Business Operating System](/business-operating-system) page.

| Area | What the system does | What stays with your people |
|---|---|---|
| Marketing | Turns customer questions and sales calls into draft posts, emails and pages, schedules them and writes the monthly report | Approving the message and the tone |
| Sales | Builds and scores prospect lists, researches each firm, drafts first emails and follow-ups, keeps customer records up to date | Approving what is sent and taking the meeting |
| Customer care | Answers common questions in minutes from your own help pages and passes hard cases to a person with a summary | Complaints and judgement calls |
| Finance | Raises the invoice when a deal is signed, chases late payers on a schedule, sends a weekly cash report | Your accountant signs off the books |
| Contracts | Fills standard contracts from your templates, tracks renewal dates, flags unusual clauses | Your solicitor gives the advice |
| Hiring | Writes job ads, books interviews, runs onboarding checklists | Your managers choose who to hire |
| Delivery | Onboards each new client step by step, sends status updates, hands work between teams | Your team does the skilled work |
| Reporting | Pulls numbers from every tool into one weekly page and flags what changed | Deciding what to do about it |

One rule holds the table together. Anything that reaches a customer or moves money is approved by a person, and every action is logged. The law points the same way: the Information Commissioner's Office says in its [draft guidance on automated decisions](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/individual-rights/automated-decision-making/what-are-the-adm-safeguards/), updated in March 2026, that people affected by a significant decision made solely by a machine must be told about it and be able to get a human to review it. For how to choose the work to start with, see [which jobs to automate first](/insights/which-jobs-to-automate-first).

## How is it different from buying software?

Software gives each person a tool. A business operating system changes who does the job. The difference shows up in the results.

Most firms are at the tools stage. The government's [AI adoption survey](https://www.gov.uk/government/publications/ai-adoption-survey-2026/ai-adoption-survey-report-2026-executive-summary), based on 3,503 interviews with UK businesses in early 2026, found that 32% were using at least one AI technology, twice the share a year earlier. Only 14% considered AI tools to be of high or critical importance to their operations, and 64% had seen no immediate financial effect.

Larger companies show the same pattern. [McKinsey's 2025 global survey](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-2025) of 1,993 respondents found 88% reporting regular AI use in at least one business function, while nearly two-thirds said their organisations had not yet begun scaling it. The small group seeing real profit from AI were nearly three times as likely as the rest to have redesigned their workflows.

That last finding is the point. A licence helps one person write an email faster. It does not raise the invoice when the deal is signed, because nobody has connected the sales record to the finance system or written down the rule. We look at the same shift from the marketing side in [whether to keep paying a marketing agency now that AI can do the work](/insights/ai-disrupting-marketing-agency-model).

## How is it different from hiring, or from a consultancy report?

Hiring adds capacity that costs the same every month and rises with every new client. A consultancy tells you what to build and leaves. A business operating system is built once, runs every month and belongs to you.

| Route | What you get | Main risk |
|---|---|---|
| Hire more people | Capacity you can see and manage, on duty in working hours | Costs rise in step with revenue, so margin stays where it was |
| Buy AI tools for the team | Faster individuals | Nothing changes in how work moves between people |
| A consultancy review | A long report and a list of recommendations | Your team still has to build it alongside their day jobs |
| Build it in-house | Full control | Trial and error on real customers, and a half-built system when the builder moves on |
| A business operating system built with a specialist | A map, a plan, then one area live at a time, run every month | It needs an owner on your side and settled processes to work from |

The evidence on who should build it is fairly clear. MIT's NANDA initiative studied 300 public AI deployments and interviewed 150 leaders. [Fortune's report of the study](https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/) says about 5% of pilots achieved rapid revenue growth and most stalled with little or no measurable effect on profit. Buying from specialist vendors and building partnerships succeeded about 67% of the time. Internal builds succeeded about a third as often.

We have seen the people-heavy version from the inside. Our founders built Venntro to $50m a year in revenue, and it took a team of 200 to run it. With the systems we build today, the same business could run with 20 people or fewer. On cost, our prices are published on [what it costs](/what-it-costs): an Operations Map and plan for one team is £2,999, a build is from £6,000 for each area, and running it is £2,500 to £5,000 a month with no minimum term. For the comparison with salaries, see [what automating a department costs compared with hiring](/insights/cost-of-automating-a-department-vs-hiring), and for marketing roles in particular, [what a fractional CMO, an agency and an in-house hire each cost](/insights/fractional-cmo-vs-agency-vs-in-house-cost).

## What does a firm of 50 to 300 people get from one?

It gets back the routine hours that are spread thinly across every team, a weekly page of numbers, and the ability to take on more work without hiring at the same rate. It does not get a smaller team overnight.

In a firm this size the routine work is not in one place. It is a few hours of everyone's week: copying, chasing, checking, re-keying. Nobody sees the total. Our example Operations Map uses a made-up commercial law firm with 120 people and £14m in fees. Adding up each team gives 290 hours of routine work a week, of which a system could take on 170. Those figures show how a map works. They are not a forecast for your firm.

Three outside findings explain why this size of firm is a good fit.

- **People are short of time, not ideas.** Microsoft's [2025 Work Trend Index](https://www.microsoft.com/en-us/worklab/work-trend-index/2025-the-year-the-frontier-firm-is-born), a survey of about 31,000 knowledge workers in 31 markets, found 80% saying they lack enough time or energy to do their work.
- **Mid-sized firms measure too little.** The ONS scores [management practices](https://www.ons.gov.uk/economy/economicoutputandproductivity/productivitymeasures/bulletins/managementpracticesintheuk/latest) from 0 to 1. Firms with 50 to 99 employees averaged 0.63 in 2023, and use of key performance indicators was the weakest area for firms overall, at 0.42. A weekly one-page dashboard goes straight at that gap.
- **The gain so far is growth without extra hires, not job cuts.** The ONS found that most businesses using AI report no change in headcount, and just under 7% of medium-sized businesses report a decrease.

We run our own firm this way. Five of us do the work that took about 30 people at our previous companies, with an extended team to call on when a client needs one. The detail, area by area, is on [how we run Fortitude](/how-we-run-fortitude).

## How do you build one without stopping the business?

One area at a time, without ripping out what you have. These are the steps we follow, and a capable in-house team can follow the same ones.

### Step 1: Map how the work really gets done

Sit with each team and write down what they do in a normal week. Count the routine hours and mark which a system could take on. For one team this takes about three weeks.

### Step 2: Agree a plan with your leadership team

Decide what to build first, with which tools, what it costs and what it should save. Your leadership team signs it off, and your operations lead sets the priorities.

### Step 3: Build one area

Start where the payback is quickest. For most of our clients that is winning new work, because it also puts every customer and conversation in one place. An area takes four to eight weeks to build, test and switch on, with your team trained.

### Step 4: Run it and improve it every month

Watch it, fix what breaks and review it with the leadership team each month.

### Step 5: Keep your people in charge

Set the approval rules before anything goes live: who signs off what, and where the log is kept. Keep your data in your own accounts, decide which AI models see what, and make sure the tools and the written playbook belong to you.

Done this way, a first area is live seven to eleven weeks after the start.

## When is it too early?

It is too early when there is nothing stable to build on, or too little routine work to justify the effort. Be honest about these five signs.

- **The process changes every month.** A system needs a settled way of doing the job. If you are still working out how to deliver your service, write the playbook first and automate later.
- **The routine work is small.** If the routine hours across the firm add up to less than one full-time role, the saving is unlikely to cover the build and the upkeep.
- **Nobody senior will own it.** Without a named owner who sets priorities and signs off each build, the project becomes another pilot.
- **Your records are in no fit state.** If customer and money data sit in personal spreadsheets, the first job is the single record, and it may be the only job for a while.
- **You want it to make the decisions.** If the aim is to remove people from complaints, advice or hiring choices, stop. Those stay with people.

## What to do this month

1. Write one sentence saying what you want from AI: more revenue with the same team, faster service, or fewer errors.
2. Ask each team lead to list the five tasks their people repeat most often, with a rough count of hours a week.
3. Check whether your customers, deals and invoices live in one place or several.
4. Pick one process and write it down as steps a new starter could follow. That is the first page of your playbook.
5. Name the person on your leadership team who would own this, and agree what will never be handed to a system.

If you would like a second pair of eyes on that list, [book a conversation](/growth-review): 30 minutes with a founder, at no charge, and you leave with three areas a system could take on, in writing.

## Frequently asked questions

### Is a business operating system the same as EOS?

No. EOS is a management framework for leadership teams, covering vision, people, data, issues, process and traction. A business operating system in our sense is the set of systems that carries out routine work. A firm can run EOS meetings and still copy data between spreadsheets by hand.

### Will it replace our people?

No. It takes the routine work off them so they can do the work that needs a person. Most firms use it to grow without adding headcount at the same rate.

### Do we have to replace our CRM or finance system?

Usually not. A sound build sits on the customer, finance and practice tools you already use wherever it can. Your data stays in your accounts, and you approve which AI models see what.

### Can we build it ourselves?

Yes. A good map and plan should say which areas suit your own team and which are better built by specialists. Bear in mind the MIT finding reported by Fortune: internal builds succeeded about a third as often as projects bought from or built with a specialist.

### How long before the first part is working?

On our timings, a first area is live seven to eleven weeks after the start: about three weeks to map and plan, then four to eight weeks to build. The rest follow one area at a time.

## Sources

- [Office for National Statistics, Artificial intelligence in UK businesses: 2023 to 2026 (July 2026)](https://www.ons.gov.uk/businessindustryandtrade/business/businessservices/articles/artificialintelligenceinukbusinesses/2023to2026)
- [GOV.UK, AI adoption survey report 2026: executive summary (October 2026)](https://www.gov.uk/government/publications/ai-adoption-survey-2026/ai-adoption-survey-report-2026-executive-summary)
- [McKinsey, The state of AI in 2025: Agents, innovation, and transformation (November 2025)](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-2025)
- [Fortune, MIT report: 95% of generative AI pilots at companies are failing (August 2025)](https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/)
- [EOS Worldwide, What is EOS and is it right for your business? (January 2026)](https://www.eosworldwide.com/blog/what-is-eos-the-entrepreneurial-operating-system-and-is-it-right-for-your-business)
- [Information Commissioner's Office, What are the ADM safeguards? Draft guidance (updated March 2026)](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/individual-rights/automated-decision-making/what-are-the-adm-safeguards/)
- [Microsoft, 2025 Work Trend Index: The year the Frontier Firm is born (April 2025)](https://www.microsoft.com/en-us/worklab/work-trend-index/2025-the-year-the-frontier-firm-is-born)
- [Office for National Statistics, Management practices in the UK: 2016 to 2023 (May 2024)](https://www.ons.gov.uk/economy/economicoutputandproductivity/productivitymeasures/bulletins/managementpracticesintheuk/latest)
