Zerodha XIRR accuracy: Why SIP returns differ slightly

Mutual fund investors using SIPs keep comparing the XIRR shown on Zerodha Coin with Excel, Google Sheets, Groww, Kuvera, and MFCentral. The discussion is less about the concept of XIRR and more about why the displayed number can look a little different across screens. XIRR is widely described as the most accurate return metric for real-life investing because it includes every cash flow and its exact date. It is also the method commonly referenced as the standard used across Indian mutual fund platforms for SIP return reporting. That standardisation is why users expect identical results when the same fund and the same transactions are tracked on different apps. Still, posts highlight that small discrepancies can show up even when platforms use the same underlying math. The practical takeaway is that the formula can be consistent while the inputs or the valuation timestamp can vary slightly. Understanding what inputs each platform uses is usually enough to reconcile the gap.

What XIRR measures for SIP investors

XIRR, or Extended Internal Rate of Return, is the annualised return that makes the net present value of all cash flows equal to zero. In plain terms, it tries to compress a long list of investments made on different dates into one yearly percentage. This matters for SIPs because each instalment is invested at a different NAV and stays invested for a different duration. CAGR is often mentioned as a two-point measure that works cleanly for a single lumpsum, but it does not fully represent staggered cash flows. XIRR handles irregular dates and multiple cash flows, which is why it is regularly recommended for SIP performance. The standard equation is typically described as summing each cash flow discounted by (1+r) raised to a day-count fraction. Because r cannot be rearranged directly, XIRR is solved iteratively, commonly using a Newton-Raphson approach. This iterative nature is also why some people see tiny differences between tools when inputs or rounding are not identical.

Why platforms default to XIRR over CAGR

The core argument repeated across threads is that XIRR is the only mathematically correct measure for SIPs. Many Indian apps show SIP returns as XIRR because it fits the cash-flow pattern of monthly or irregular investing. Social posts describe it as the standard method used across major platforms, including Zerodha Coin and Groww, for mutual fund SIP reporting. The emphasis is that XIRR factors in pauses, additional investments, partial redemptions, and any other personal timeline differences. That leads to a key distinction often missed by new investors: the fund’s “standard” SIP return is different from an investor’s personal XIRR. Your personal XIRR depends on your actual dates and amounts, not a model SIP schedule. If you skipped a few months or topped up after a bonus, your number will differ from factsheet SIP return figures. As a result, “my XIRR looks wrong” complaints are sometimes just mismatched comparisons between different timelines.

Zerodha Coin vs Groww: same math, not always same inputs

A recurring claim in the discussion is that Zerodha and Groww use the same mathematical XIRR formula. Users also point out that the transaction and NAV data ultimately comes from the same registrar ecosystem, which sets expectations of consistency. When people do see a difference, the examples described are typically very small, such as a few basis points. The most cited reason is valuation timing, such as when an app fetches the end-of-day NAV or refreshes holdings. Another reason mentioned is the way platforms handle fractional units, which can change the last decimal in units and therefore the current value. Dividend reinvestments are also cited as an area where treatment can differ in presentation or timing, even if the overall method is still XIRR. Since XIRR uses dates, even a one-day shift in the final valuation date can move the annualised number slightly. The important point from the thread is that small differences do not automatically imply “wrong calculation”, they often imply “slightly different final cash flow or valuation snapshot”.

The most common causes of SIP XIRR discrepancy

Most discrepancies discussed online come down to how cash flows are captured. XIRR requires every SIP instalment as a negative cash flow, and the current value or redemption as a positive cash flow with a date. If any instalment date is off, or if the sign convention is reversed, the result can swing meaningfully. Another common issue is missing transactions when a person checks XIRR only inside one platform, while investments exist on multiple platforms. Threads also highlight that some apps can show portfolio return only for investments made through that app, not an investor’s full mutual fund universe. Rounding and fractional units can cause small basis-point differences even when transactions match. Timing of NAV updates can also cause micro gaps between two apps viewed at different times of the day. Finally, iterative methods can converge slightly differently across tools depending on guess values and numerical tolerances, which is why some users cite matches “within about 0.1%” rather than perfect equality.

Quick reference table: XIRR vs CAGR for SIP tracking

The comparison most repeated in posts is that CAGR is simpler but mismatched for SIPs, while XIRR is designed for multi-date cash flows. This is why people are advised to use XIRR for SIP portfolios running longer than a year. The key is not the label on the app screen, but whether the method uses every investment date and amount. The table below summarises the practical differences highlighted in the discussion.

How to verify Zerodha Coin XIRR in Excel or Sheets

The social walkthrough is consistent: list dates in one column and cash flows in another. Each SIP payment should be entered as a negative number because money is going out of your pocket. Add the current portfolio value as a positive number on the last row with today’s date. Then use the XIRR function by selecting the values range and the date range, optionally adding a guess such as 0.1. People stress formatting and completeness, because a single missing SIP entry can distort the result. If you redeemed partially, that redemption should be a positive cash flow on the redemption date. If you switched funds or paused SIPs, those gaps should naturally appear as missing cash flows, not as an averaged assumption. When this is done cleanly, users report that platform and Excel results usually align closely, with any remaining gap often traceable to valuation timing or rounding.

Why your “personal XIRR” can differ from factsheet SIP returns

Another frequent source of confusion is comparing personal XIRR with the fund’s published SIP return numbers. Factsheet SIP returns, when available, reflect a defined SIP schedule and a standard period, not your exact transaction history. Your app’s XIRR is based on what you actually did - the exact dates you paid SIPs, any pauses, and any extra purchases. If you invested through different routes, like some instalments through one app and lumpsums elsewhere, each platform can only compute XIRR on the slice it can see. That is why a user can see one XIRR on Zerodha Coin and a different one on another platform, even with the same fund name. Posts also note that to get a complete view across platforms, investors often rely on a consolidated account statement and compute XIRR themselves. The principle is straightforward: XIRR is only as complete as the cash-flow list. So, “fund XIRR”, “platform XIRR”, and “portfolio-wide XIRR” can all be different numbers for valid reasons.

Pre-tax numbers and what XIRR does not include

The discussion also flags that app-reported XIRR is a pre-tax, gross return. It does not account for capital gains tax that becomes payable when you redeem units, so realised post-tax returns will differ. This can matter when investors compare XIRR with what they actually take home after a redemption. Another limitation raised is that broker-style portfolio XIRR can be incomplete if it is not built from full cash-flow records, especially when people think in terms of cash deposited versus investment purchases. Some posts argue that idle cash timing is not always reflected in broker-return displays, which can make numbers look better than reality. For mutual funds specifically, the cleanest approach is still to rely on the mutual fund cash flows and the current valuation as of a specific date. If you want a single number for everything, you need every cash flow across every platform. In other words, XIRR is accurate as a method, but the displayed result depends on whether the dataset is complete and consistently timestamped.

A practical checklist to reconcile small gaps

Start by confirming that every SIP instalment appears once, with the correct date and a negative sign in your own sheet. Next, confirm the final positive cash flow equals the current value you are comparing against, and that the date matches the date on the app’s valuation. If you are comparing across apps, check whether both are showing the same holding type and the same platform-only slice of your investments. Then look for events like dividend reinvestments that could introduce additional dated cash flows. If the gap is only a few basis points, consider the timing explanation discussed online, such as end-of-day NAV refresh differences. Also remember that XIRR is solved iteratively, so two tools can converge to slightly different decimals even with the same method. Finally, if you want a portfolio-wide number across apps, pull a consolidated statement and compute a single XIRR from that combined cash-flow list. This approach usually turns “Zerodha XIRR is wrong” into a concrete, fixable input mismatch.