Site Expense Register Format: How to Track Site Kharcha Without Fights

Every contractor knows this scene. The month ends, the client asks for a summary, and the munshi brings a notebook. Some entries have amounts. Some have only a name. Half the bills are missing. You ask what Rs 8,000 went for on the 14th, and nobody can say for sure. This is not a case of theft, most of the time. It is a case of no system. A site expense register, kept properly, removes most of these fights. This article gives you a simple format, a petty cash method, and a weekly check that takes ten minutes.

What counts as a site expense

Before you make a register, agree on what goes into it. If every person on site has a different idea, the register will still cause fights. A site expense is any payment made for a specific site, small or large, that is not a formal purchase order against a vendor bill. Common examples:

• Miscellaneous material bought in a hurry: nails, wire, cement one bag at a time, cutting wheels.
• Transport: tempo hire, loading and unloading charges, auto fare for carrying material.
• Tool repair and small tools: drill bits, blades, machine servicing.
• Labour payments outside the contractor’s contract: extra helpers, night shift tea and food.
• Site running costs: water, electricity, tea for workers, safety items like helmets and gloves.
• Fuel for generator or machines.

What does NOT belong in this register: scheduled material orders against a purchase order (those go in your material register), labour attendance-based payments (those go in your labour and wages records), and big vendor bills paid by cheque or bank transfer against a formal bill. Keeping these separate makes each register simpler and easier to check. If you are also struggling with material records, see our material management approach.

The register format

Use the same columns whether you keep it in a notebook, an Excel sheet, or an app. Below is the format we recommend. You can also start with our free construction Excel templates if you want a ready sheet.

Two rules make this format work. First, no entry without at least a photo of the bill. Second, the person who spends and the person who approves must be different people. Even in a small firm with three staff, this separation is possible and it changes behaviour.

If you use Excel for other site records too, you may also want a material stock register format so both registers follow the same pattern.

Petty cash with a site person: a simple imprest method

The most common site kharcha problem is not the register. It is the open cash box. Someone is given cash “as needed”, spends it, and the balance never matches. The fix is an imprest system. It is old, boring, and it works.

How the imprest system works

1. Fix a float amount per site. Say Rs 5,000 for a small residential site.
2. Give Rs 5,000 to the site person. This is their opening balance.
3. They spend and record each item in the register with a bill or photo.
4. At a fixed day, say every Saturday, you count the cash left and add up the bills. Cash in hand plus bills should equal Rs 5,000.
5. If it matches, you reimburse exactly what was spent, bringing the box back to Rs 5,000.
6. If it does not match, that gap is discussed before the next float is given. No reimbursement until it is settled.

Example: On Monday the site person has Rs 5,000. During the week they spend Rs 1,200 on tempo hire, Rs 650 on misc material, and Rs 300 on worker tea. Total spent Rs 2,150. On Saturday the box should hold Rs 2,850 in cash plus bills and photos worth Rs 2,150. If the box holds only Rs 2,600, there is Rs 250 unexplained. Small amount, but catching Rs 250 every week is what prevents Rs 25,000 going missing over a year.

The key point: the person never gets fresh money “against a note”. They get money only against bills and the counted balance. This one rule removes most arguments, because the maths is public and simple.

A ten minute weekly check for the owner

A register nobody checks is worthless within a month. You do not need daily checking. Ten minutes on a fixed day is enough. Do this:

• Count the petty cash box and check cash plus bills equals the float. Ten seconds.
• Scan the week’s entries. Look for entries with no bill, round amounts appearing too often, and the “other” category being used more than once or twice. Two minutes.
• Pick two entries at random and ask to see the bill photo. Call the shop or tempo person if something looks odd. Two minutes.
• Check category totals against your rough expectation. If transport is Rs 4,000 in a week where the tempo came twice, ask. Two minutes.
• Sign or mark the last checked row. This tells everyone the register was reviewed up to that point. One minute.
• Note anything pending for month end, like a bill promised by a shop. Two minutes.

The purpose of this check is not to catch anyone. It is that everyone knows the check happens. That knowledge alone keeps the register clean. If you want to see how a digital version of this looks, our reports and dashboards page shows the idea, and a WhatsApp demo takes a few minutes.

Where a register breaks, and site-wise payments in Trackovo

A paper or Excel register works well up to two or three sites. Beyond that, it breaks in predictable ways:

• Entries are made two or three days late, from memory. Memory is not a receipt.
• The Excel file lives on one laptop. The owner cannot see it from home, so checking slips.
• Bill photos sit in somebody’s phone gallery mixed with personal photos, and get deleted.
• Site-wise totals need manual sorting every month, and every sort is a chance for a formula error.
• There is no clean link between a site expense and the client bill or the project budget, so cost per site is only known late.

This is where an app earns its place. In Trackovo’s finance management, site-wise payments and expenses are recorded from the phone at the site itself, with the bill photo attached to the entry. The entered-by and approved-by steps are built in, so the same separation of duties from the paper register exists but cannot be skipped. Totals by site, by category, and by month come without any sorting, and the same app also handles attendance and salary, so extra labour payments and attendance-based wages stay connected instead of living in two notebooks.

To be fair, we are not the only option. Powerplay and OnSite are construction management apps used by Indian contractors, and both are worth a look. The honest way to compare is to list what you actually need: number of sites, whether you need expense approval, whether you need the same app to handle hajiri and material, and your monthly budget. Then check each app’s own website for current pricing and features, or ask each vendor for a written quote. We have written our own comparisons here: Trackovo vs Powerplay and Trackovo vs OnSite. Our plans are listed openly: Starter at Rs 1,799 per month, or Rs 17,999 per year, for 3 core users and 5 active projects, and Growth at Rs 2,999 per month or Rs 29,999 per year for 6 core users and 12 projects; Scale, for larger builders, is priced on request. GST at 18 percent is extra. Every module is in every plan, and there is a 14-day free trial with no card needed.

If attendance is your bigger headache right now, start with our guides on the labour attendance sheet format in Excel, the hajiri app for construction labour, and the best labour attendance apps for construction in India in 2026. For a wider view of software choices, see our piece on the best construction management software in India in 2026. You can also explore everything in one place on the features page or book a demo.

Common questions

Should I keep one register for all sites or one per site?

For up to three sites, one register with a site column is fine and easier to check. Beyond that, one per site works better because each site person records in their own register and the numbers stay separate. In an app this is handled automatically: you record on your phone, tag the site, and the site-wise totals are separate without any extra effort. The format of the columns stays the same either way.

What if the shop refuses to give a bill for small amounts?

This is common for tiny purchases. The practical rule is: no bill, then a photo. Ask for any rough slip, or photograph the material with the amount written on paper next to it, or photograph the UPI payment screen. It is not a tax invoice and should not be treated as one, but it proves the purchase happened. What you must not accept is an entry with neither a bill nor a photo, because that is exactly the entry that causes the month-end fight.

How much petty cash float is right for a site?

There is no fixed number, but a useful guide is one week of expected misc spending plus a small buffer. For a small residential interior site this might be Rs 3,000 to Rs 5,000. For a larger structure with generators and daily transport, it will be higher. Keep it as low as practical, because a small float means small mistakes, and reconcile it weekly on a fixed day so the amount never grows silently. If a site regularly runs out mid-week, increase the float at the next reconciliation rather than giving ad hoc top-ups.

Can I run this whole thing in Excel instead of an app?

Yes, for a small operation Excel works, and many contractors run it for years. Use the same columns given above, one sheet per month, and a bill photo folder named by date. The honest limitation is discipline: entries get delayed, photos sit in phones, and the owner cannot check remotely. An app mainly removes those friction points and gives you ready site-wise totals and daily reports, like a DPR built from the same data. Start with Excel if that is your comfort level, and move to an app when checking starts slipping or you cross three or four active sites.