A Durga Puja budget is unusual: the biggest commitments — the idol, the pandal, the artist — are made months before most of the chanda is collected. You are spending against money you do not have yet. That is manageable, but only if you know two numbers at any moment: what you have actually collected, and what you have actually committed.
Start from last year, not from zero
The most accurate predictor of this year’s collection is last year’s collection, adjusted for how many units or households have joined or left. If your previous years live in a diary, you are guessing. If they are in one place, you are planning.
The cost lines that actually move
Every Puja is different, but the budget almost always breaks down along the same lines. Track them as categories from day one so that next year you can compare like with like.
- Pratima — the idol, plus transport and immersion
- Pandal — structure, decoration, lighting
- Purohit, bhog and pooja samagri
- Dhak, sound and cultural programme
- Security, permits and municipal charges
- Prasad and distribution
Advances are where the money leaks
The pandal decorator takes 40% up front. The idol artist takes a booking amount in Ashadh. Somebody hands ₹15,000 in cash to a volunteer to buy samagri. Each one of these is real money out, and none of them look like an “expense” yet — which is exactly why they go missing from the reckoning.
Record the advance the day it leaves your hands, against the vendor or the committee member who took it, and record the balance when the final bill lands. Kosh tracks vendor deals and part-payments so the balance due is always visible, and committee advances are settled against the bills that member actually produced.
Publish an itemised account, not a total
“We collected ₹8,40,000 and spent ₹8,15,000” tells your subscribers nothing and invites suspicion. An itemised account — every head, every vendor, every bill — tells them everything and ends the conversation. It is the same information, and it takes one tap when the records were kept as you went.