The AI Conversation You Have Been Quietly Avoiding
Everyone says you should be using AI. Nobody says what for. So you have done nothing, and you feel behind. Here is the honest version.
Practical writing on the systems behind growth: operations, revenue, pricing, marketing and applying AI inside a working business.
Everyone says you should be using AI. Nobody says what for. So you have done nothing, and you feel behind. Here is the honest version.
Nobody cancelled. Nobody complained. They just stopped coming, and you only noticed months later when the numbers flattened.
The customer asked on the 3rd. You sent it on the 17th. They had already decided by the 8th, and you never found out why you lost.
You have been trading for eleven years and doing excellent work. Search for what you do in your own city and you are nowhere. Both are true at once.
You spent a day on strategy. Everyone left energised. Three months later almost none of it has happened, and nobody has mentioned it since.
Your clinical work is excellent. Your reviews say so. And on Tuesday at three, the waiting room is empty and you are doing the arithmetic in your head.
Every table is taken on a Saturday. The kitchen is flat out. And at the end of the month there is nothing left. This is more common than you think.
You spend on ads every month. The enquiries arrive. And somewhere between arriving and being called, a large share of them simply stop existing.
It worked for twenty years because everyone knew everything. Now there are forty people, and the way it always worked has stopped working.
They are cheaper, their work is not as good, and customers keep choosing them. It is maddening, and there is a reason for it.
You told yourself the market slowed down. Then you looked at the enquiry numbers and realised the market did exactly what it always does.
You have explained it four times. They are not being careless, and you are not being unclear. Something else is going on.
You paid for it, you like it, people compliment it. And it has generated four enquiries this year. Both things can be true.
You know the number is too low. You have known for a year. Here is what is actually stopping you, and why the fear is misplaced.
From outside it looks like everything is working. Inside, you are holding it together with effort and hoping nothing breaks this week.
The team is stretched, the orders keep coming, and somehow there is less money than last year. You are not imagining it, and you are not bad at this.
Nobody complains about the call that was never returned. That is exactly why it keeps happening, and why it never shows up in any report.
You have not taken a proper holiday in three years. Every decision routes through you. You built this, and somewhere along the way it started owning you.
They are your largest account, everyone knows their name, and you would be quietly relieved if they left. Here is why that feeling is data.
Your team works hard. You can see it. So why does everything take three weeks, and why does asking about it make everyone defensive?
Revenue leakage rarely announces itself. It hides in unanswered enquiries, unbilled work, and discounts nobody approved. Here is where to look, in order.
Most SOP projects fail the same way: a document is written, filed, and never opened again. The problem is rarely the writing. It is where the SOP lives.
When conversion drops, most businesses buy more leads. The data usually says the leads were never the constraint — the handling was.
Busy and profitable are different conditions. When revenue climbs and margin does not, pricing is usually the variable nobody revisited.
Before changing anything, find out what is actually true. Thirty questions across six areas, answerable in a week with data you already hold.
Customers now ask ChatGPT, Perplexity and Google AI Overviews for recommendations. Being cited in those answers is a different discipline from ranking on page one.
AI pays back fastest on high-volume, low-judgement, repetitive work. Applied anywhere else it usually adds a system to maintain and little else.
Reach, impressions and followers can all rise while enquiries stay flat. The gap is almost always positioning and qualification, not budget.
Most businesses have a defined process for winning a customer and nothing at all for keeping one. That asymmetry is where lifetime value disappears.
Adding people to a broken process buys relief for a quarter and adds cost permanently. Here is how to tell which problem you actually have.
Patients choose on trust and convenience long before they can judge clinical quality. Most clinic growth problems are business-system problems.
A busy restaurant can be an unprofitable one. The leaks are in pricing structure, waste, and inconsistency — not in the number of covers.
Buying more leads when conversion is broken multiplies cost, not sales. Qualification and follow-up discipline decide real estate outcomes.
Overstock ties up cash and ends in discount. Understock loses sales you already paid to create. Both are forecasting problems.
Premature process kills speed. No process kills the company at the moment it starts working. Here is the line between them.
Growth consumes cash before it produces it. Businesses fail while profitable more often than most owners expect.
Technical capability rarely limits an industrial business. Coordination, pricing discipline and sales structure usually do.
Education businesses rarely lose students on quality of teaching. They lose them in the days between first enquiry and enrolment.
Consulting engagements fail on scope and accountability far more often than on expertise. Here is what to establish before you commit.
Without knowing the cost to acquire and the profit to serve one customer, every growth decision is a guess dressed as strategy.
Big decisions are usually made once, on one set of assumptions. Scenario planning tests them cheaply before capital is committed.
Resilience is not caution. It is structural flexibility — the ability to absorb a shock without the business stopping.
Most businesses do not slow down because decisions are hard. They slow down because nobody knows who makes them.
Direct-to-customer removes the intermediary but adds operations. Farms that plan for that keep the margin; farms that do not lose it to logistics.
You have been told you need a consultant. Nobody has told you what actually happens, what it costs you in time, or how to tell a good one from an expensive one.
The first branch ran on you knowing everyone. The second one has the same equipment, the same prices, and a completely different patient experience.
Expansion rarely fails because the new location is wrong. It fails because the original store was held together by habits nobody ever wrote down.
Ninety percent occupancy feels like success. If you got there by discounting, you have simply sold the same rooms to a worse customer for less money.
You spent months and a great deal of money winning that customer. Then you handed over the keys and let the most valuable relationship in your business go cold.
Winning more quotes is not always good news. If your costing is optimistic, every extra order you win takes money off the table a little faster.
Most farm businesses price against what the market is paying. Very few can say what a kilo actually costs them to produce, which makes every deal a guess.
Admissions get all the attention and all the budget. The students who drift away mid-course cost more, and almost nobody is watching for them.
Your calendar has no gaps and your team is stretched. If the profit does not reflect that, you are selling capacity you have never actually measured.
Founder-led sales works beautifully until it becomes the ceiling. Handing it over feels like a downgrade because, for the first few months, it genuinely is.
Most Indian businesses now receive their best leads on WhatsApp. Very few treat that inbox as a sales channel, so the fastest competitor wins by default.
Every vendor demo is impressive and every trial starts well. Businesses that get value from AI define the problem before they look at a single tool.
They are not better than you and they did not spend more. They filled in the parts of their listing you left blank, and they did it consistently.
If your competitor could put their name on your website and nothing would look wrong, you do not have positioning. You have a description.
Chasing payments feels like begging for something already yours. That discomfort is why most businesses do it late, badly, and only when cash gets tight.
It started as an exception for one important customer. Two years later it is what everybody expects, and nobody can remember deciding that.
They are late, the quality varies, and switching feels risky. So you absorb the cost quarter after quarter and call it the way the industry works.
You have delegated the same task three times and it has returned every time. The problem is usually not the person, and it is rarely their ability.
The business has grown past the point where consensus at dinner is a decision-making system. Nobody wants to say so, so decisions simply stop happening.
The new city looked obvious: bigger population, visible demand, a competitor doing well there. Eighteen months on it still consumes cash and attention.
Subscribe to our bespoke newsletter for monthly digests of our most critical institutional insights.
The blog covers strategy, operations, digital transformation, and building institutional systems that scale — written for business owners and operators rather than for a consulting audience.
Articles are written by the DOMS Global LLP team, drawing on the operational, revenue, and pricing patterns seen across client engagements in different industries.
New articles are added as engagement work produces patterns worth documenting. To be notified, subscribe using the form in the site footer.