Markazi Panel

What Google Analytics does not tell a store owner

Google Analytics measures what happened on your website. It has no way of knowing what your goods cost, what you paid the courier, which parcels came back, or how much of the order value actually reached your bank account. So the revenue figure it reports is accurate and almost useless on its own: it is the top of the sum, and every deduction that decides whether the month worked happens somewhere Analytics cannot see.

Analytics is a traffic tool that learned to count money

That is not a criticism, it is the design. Analytics was built to tell you where visitors came from and what they did. Ecommerce reporting was added on top, and it works by being told the value of an order at the moment the order is placed. That single number — order value at checkout — is the entire financial picture it has.

For a business that takes card payments, ships one product class at a known margin and rarely sees a return, that number is close enough to the truth to steer by. For a store in Pakistan selling cash on delivery through a courier, it is not close at all.

The four deductions it cannot see

1. What the goods cost you

Analytics does not know your cost price, and there is nowhere sensible to tell it. Two orders of Rs 5,000 look identical in every report, whether one cost you Rs 1,200 to fulfil and the other Rs 4,100. Every conclusion you draw about which product to push, which campaign to fund and which customer is worth keeping is drawn from a number that has your buying price stripped out of it.

2. Delivery

Couriers price by zone, weight and service, so the cost of getting a parcel to a customer is not one figure and is not knowable from the order. It is also charged whether or not the parcel is accepted, which matters more than it sounds — see the next one.

3. The parcels that come back

A refused COD parcel is the expensive kind of failure. You paid to send it, you pay again to get it back, the packaging is usually spent, and the stock returns to the shelf days or weeks later in whatever condition it survived. In Analytics that order was a completed purchase and stays one for ever.

This is the deduction that does the most damage, because it is not spread evenly. Returns cluster by channel, by city and by product. Two campaigns that each brought Rs 300,000 of orders are not equal if one of them brought a third of it back, and no report that counts gross orders will ever show you the difference. It is also the reason a return on ad spend calculated from gross revenue flatters exactly the campaign you should be cutting.

4. What the courier actually remitted

With cash on delivery, the sale and the money are separate events weeks apart. The customer pays the rider; the courier holds the cash, deducts its charges and a government withholding deduction, and remits the balance on its own cycle. What lands in your account is therefore never the order value, and reconciling the two is a job somebody has to do by hand. Analytics booked the full amount the day the order was placed.

Why the gap is invisible rather than obvious

Each deduction is individually small enough to feel like a rounding error and collectively large enough to be the whole margin. None of them appears on the dashboard people actually look at. And the number that is on that dashboard is real — nobody is lying to you, which is precisely why it goes unquestioned for months.

The tell is a familiar one: revenue climbing while the bank balance does not. If that sentence describes your last quarter, the problem is almost never that the revenue figure is wrong. It is that revenue was never the figure to watch.

What to measure instead

You do not need a new tool to start. You need four columns next to every order, and the discipline to fill them in:

  • Cost of goods, per variant rather than per product — sizes and packs rarely cost the same.
  • Delivery actually charged, taken from the courier’s bill, not from a flat rate you assumed at checkout.
  • Status, kept honest — specifically, whether the parcel was delivered, refused or is still out. An order that never arrives is not a sale.
  • Amount remitted, matched to the orders it settles, so a short payment is visible as a gap rather than absorbed into a monthly total.

Do that in a spreadsheet and you will already know more than the analytics does. The reason most owners stop is that the fourth column is genuinely tedious — remittances arrive as a list of consignment numbers that has to be matched against your own orders, one courier at a time, and the moment you fall a fortnight behind it stops being worth catching up.

Keep the analytics

None of this argues for removing Google Analytics. It answers a question no financial report answers — where people came from and what they did before they bought — and that question still matters. The mistake is treating its revenue line as the score. Traffic tools measure traffic. What you keep is a different number and it has to be assembled from somewhere else.

The next guide walks through a single COD order from placement to remittance, with every deduction in the order it happens.

Written by Markazi Panel, which does the arithmetic in this guide for you — see a dashboard with a month of data in it, or ask anything at hello@markazipanel.com.

Read next