Billable hours that become actual margin.

A 60-person development firm billed about ₹8 crore last year. Every project manager can show that their projects made money. The rate cards are sensible, salaries are in line with the market, and the biggest clients are well known. Yet the year closed at a single-digit operating margin, and the founder is paying salaries some months out of the previous month's collections.

The money didn't disappear in one place. Some went to people sitting between projects for three weeks at a time. Some went to fixed-price projects that ran 30% over the hours they were quoted on. Some went to change requests that were delivered and never billed. And a good share of it is still sitting with two large clients who pay on day 75 of 45-day terms.

Services businesses are simple to describe and hard to run well financially, because the product is people's time, and time that isn't billed can't be stored and sold later. The finance work that matters most is making that visible, project by project and person by person, early enough to change it.

Utilisation is billed hours as a share of available hours. Realisation is what you actually invoice and collect as a share of what those hours are worth at your rate card, after discounts, write-offs and unbilled work. Between them they explain most of the gap between a good-looking rate card and a disappointing P&L.

Same people, same rate card, same salaries. Eight points of utilisation and five of realisation add about ₹12.3 lakh of delivery margin a month, close to ₹1.5 crore a year. No price rise or new client could have achieved that as cheaply.

The hard part is that neither number is visible without timesheets that people actually fill in and a habit of comparing hours logged against hours billed on every project, every month. Most firms have the first in some form. Few have the second.

Some bench is necessary. A firm with nobody available can't start a new project quickly, and a firm that hires only after a contract is signed loses deals to competitors who can staff them next week. The mistake isn't having a bench. It's not knowing what it costs or deciding how big it should be.

An illustrative way to think about it: if six people in a 60-person firm are unbilled for a month at a fully loaded cost of ₹1.15 lakh each, the bench costs ₹6.9 lakh that month. If that's a deliberate buffer that wins two projects a quarter, it may be the best money the firm spends. If it's the leftover from a project that ended early and a sales pipeline nobody forecast, it's a planning gap wearing the label of flexibility.

Track bench cost monthly, set a target range, and connect hiring to the pipeline with a stated conversion assumption. That one link, between sales forecast and hiring plan, is where services firms most often lose money in a growth year.

Domestic clients will usually deduct TDS from your invoices. Under what is now section 393 of the Income Tax Act 2025, formerly 194J, the rate is commonly 10% for professional services and 2% for technical services, above an annual threshold of ₹50,000 per client. Receipts arrive net, and the deducted amount only comes back if it's reconciled against what clients report and claimed. Firms that don't reconcile it regularly often find credits missing at year end.

If you're registered as a micro or small enterprise, the MSMED Act gives you a real tool against slow payers. Buyers must pay within 15 days where there's no written agreement, and within 45 days at most where there is one. Late payment carries compound interest at three times the bank rate notified by the RBI. The income tax rule that denies buyers a deduction for such dues until they're paid, previously section 43B(h), carries into section 37 of the Income Tax Act 2025 from the 2026-27 financial year. Large clients know about it, and many now ask suppliers to confirm their MSME status at onboarding.

For clients outside India, the export of services rules apply: zero-rated for GST when the conditions are met, including payment in convertible foreign exchange, usually under a Letter of Undertaking filed for each financial year. Accumulated input credit then becomes a refund claim, which is cash that arrives slowly.

It rarely happens by decision. A good client keeps adding scope, the team gets comfortable, and one day that account is 38% of revenue. The problem isn't that the client is bad. It's that the firm's hiring, office and overheads are now sized to an income stream that one procurement team can reduce with a single email.

Concentration also changes negotiating power in ways that show up in the numbers. The largest clients tend to push for longer payment terms, annual rate freezes and extra unbilled effort, and a firm that depends on them tends to agree. Realisation and receivable days for the biggest account are often the worst in the business.

There's no correct limit. What helps is putting the number in the monthly MIS, modelling what the firm would look like if that client halved its spend with three months' notice, and deciding in advance which costs would change. Firms that have done that exercise make calmer decisions when the email eventually arrives.

Salaries in Indian tech rise every year. Client rates, for many firms, don't. Three years of annual increments against a flat rate card is enough to take a comfortable margin down to a thin one without a single bad project.

The firms that hold margin write an annual rate review into their contracts, raise it with clients before appraisal season rather than after, and bring numbers to the conversation: what the team has delivered, how the market for the relevant skills has moved, and what the new rate is. It's an easier conversation than most founders expect, and a much harder one after the increments have already gone out.

• ·Utilisation by team and seniority, against target.
• ·Realisation by project and client, including write-offs and unbilled change requests.
• ·Delivery margin by project and by client, after fully loaded people cost.
• ·Bench cost in rupees, not just headcount.
• ·Revenue per billable person, and per total person.
• ·Receivable days by client, with anything past terms named.
• ·Share of revenue from the top one and top three clients.
• ·Attrition in delivery roles and the cost of replacing people mid-project.
• ·Pipeline coverage against the hiring plan for the next quarter.

The work shapes itself around how services firms earn. A unit economics review by project, client and team that puts utilisation and realisation next to delivery cost. An operating plan where hiring follows a weighted pipeline. A 13-week cash forecast built on how each client really pays, net of TDS. Monthly MIS that shows margin by client, not only revenue. Tax filings and contracts stay with your CA and lawyers, and Simplify works alongside them.

Many services firms are bootstrapped and intend to stay that way. The finance work is the same, and arguably matters more, because the cash in the bank is the only cushion.