Revision History

Public launch of the InvestApps.in suite front door.

• One account and one login across every InvestApps tool (shared single sign-on).
• Plan & Pricing: one subscription — ₹2,000 / year across all three apps — with a 7-day free trial (no card, no auto-renewal, nothing stored).
• Learn hub, FAQ, and the suite's Terms, Privacy, and Refund & Cancellation policies.
• Shared User Settings and emailed payment receipts.

• A goal sent from the Investment Tracker arrives with its corpus, monthly SIP, target and horizon already filled in — ready to tune.
• The Tracker measures one blended return and volatility for a goal; the planner models equity, gold and debt separately. Rather than guess, the planner rescales its own return assumptions to match the blend you are actually invested at, so its house view stays the source of truth. The banner says what it matched — the match holds at today's allocation and drifts as your glide path de-risks.
• Save back to Tracker returns the tuned target, horizon, SIP and glide path to the Tracker's goal form for confirmation.
• The Tracker counts REITs and InvITs as alternates; here they are treated as equity, because property trusts fall when shares fall. You are told when that applies to your goal.
• The handoff rides in the part of the link browsers never send to a server, and expires after an hour — nothing is uploaded.

• Choose your currency. The planner now works in any of 13 currencies — Indian Rupee, US Dollar, Euro, Pound, UAE Dirham, Singapore Dollar, Australian Dollar, Canadian Dollar, Swiss Franc, Japanese Yen, Malaysian Ringgit, Qatari Riyal and Saudi Riyal — chosen right next to the goal amount. Rupee remains the default.
• When you switch, amounts are converted at the live exchange rate, so a plan keeps meaning the same thing rather than just changing its symbol. The rate in use and how recently it was fetched are shown on screen, and the planner keeps working offline on the last rate it saw.
• Figures follow the convention of the currency you pick — lakh and crore for rupees, thousands and millions elsewhere — in the summary, the year-by-year table and the chart axes alike.
• Saved goals remember the currency they were entered in, so opening a goal while the planner is set to a different currency converts it instead of misreading it.
• New versions now announce themselves. When an update ships, the planner offers to load it instead of leaving you on the old version until you happened to force a refresh. Declining is safe — the app carries on working and asks again next time.
• Fixed: the probability panel could sit on “Running Monte Carlo…” indefinitely after a new version had shipped, instead of saying a refresh was needed.
• The outcome-distribution chart is taller, and its Goal marker is no longer clipped at the top.

• New All apps menu in the header. It lists the other InvestApps tools directly and marks the one you are in, so switching is a single click — the old link only took you to the front page to find your own way from there.
• On a phone the menu becomes an icon and the header no longer wraps onto two lines.

Public launch. Free, privacy-first goal planning.

• Set a target amount and a horizon; get the SIP required to reach it, with an optional annual step-up.
• Inflation-adjusted targets so the goal is stated in real terms.
• Everything runs in your browser — nothing is uploaded.

⚠️ Your numbers will change

• The Monte Carlo was modelling a slower portfolio than the projection above it. Every rate you type is a compound annual rate, but the simulation treated it as a simple average — so volatility was charged twice, and a 15% ± 26% sleeve actually ran near 12%. Success rates and median outcomes now read higher, and the median simulated outcome lands on the deterministic projection instead of far below one no path could reach. On one test plan the success rate went from about 85% to about 93%.
• A simulated year could lose more than everything. At high volatility roughly 1 in 10 long paths wiped out a sleeve outright. Yearly losses are now floored.
• “Final zero” in the histogram was captioned as the failure rate. It is not — a plan that ran dry and was later refilled by a deferred pension has failed but ends with money. Ran-dry and later-refilled are now reported separately.

⚠️ Custom bucket ladders

• Deleting a rung left its share behind, so shares could add up to more than 100% and the whole ladder was over-funded. Assets tagged to the deleted rung were silently re-filed into the last rung, with the Bucket dropdown showing nothing. Both are cleaned up now — a custom ladder's corpus at life expectancy may read lower.
• The de-risking roll-down froze partway and stayed frozen for the years that bucket funded your spending: a rung set to 30% equity ran nearer 13%. It now finishes the glide.
• Pinning a bucket's equity weight below what its volatility target implies made the engine derive an impossible return for the rest of the bucket. It is now held to the limits the editor itself accepts, and the feasibility check names the return your target would have needed. Ladders that let the weight follow the volatility target — the default — are unchanged.
• Clearing every bucket-size box read as “zero everywhere”, showing an intact corpus as empty and depleted in year one.

⚠️ If you use Advanced Mode with a Tax Drag above 0%, your corpus will now read lower — the old figure was too high

• The ⚙ Parameters → Tax Drag sliders exist to reduce your return assumptions to after-tax figures. In Basic Mode they did exactly that. In Advanced Mode they had no effect at all on the years before retirement.
• In Advanced Mode each asset grows at its own return, set per asset in Edit Profile → Assets — and that rate was being used gross, with the tax never taken off. The same tax was still being applied to those same assets after retirement, so a single plan was being projected on two different sets of assumptions.
• To put a size on it: moving all three sliders from 0% to 1% reduced a Basic-mode plan's corpus at retirement by about 5%. In Advanced Mode, the identical change moved it by 0%.
• The same omission applied to the Advanced SIP, whose blended return is built from your per-fund figures, and to the value used to price an NPS-style annuity at its unlock age.

Who is affected

• Advanced Mode with any Tax Drag above 0% — affected. Expect a few per cent off your corpus at retirement, and more by life expectancy — often into double digits, because a small annual drag compounds across the whole build-up and the whole of retirement. This is the correction, not a new shortfall: the plan was previously projecting returns you would not have kept after tax.
• Tax Drag left at 0%, the default — not affected. Nothing moves.
• Basic Mode — not affected. It always applied the drag correctly.
• Real estate is deliberately unchanged. There is no Real Estate Tax Drag slider, so property keeps its gross rate in both modes.

Three figures on the Buckets tab described a plan we had not run

• The bucket flow diagram said Bucket 2 is refilled once it drops below a 2-year reserve. The projection has used 3 years for some time, and in Advanced Mode the Strategy Lab lets you set it to anything — so the diagram could contradict the "B2 reserve floor" control on the same screen. It now reads the live value.
• Bucket 3's "At retirement" figure was worked out differently from the Growth · B3 card at the top of the same tab. If you route gold to Bucket 3, the two disagreed by the whole of that gold. Both now show the balance the projection actually opens with.
• The percentages captioning the three bucket cards showed your ⚙ Parameters targets. In Advanced Mode without de-risking, the split is re-derived from the buckets you tagged your assets with — so the caption and the balance printed directly beneath it could describe two different portfolios. They now both show the split the projection ran.

⚠️ If you use the 5-Bucket strategy, your numbers will change — they were wrong before

• On the dashboard, the 5-Bucket strategy read every asset's bucket from the wrong place. Edit Profile → Assets showed your assignment correctly, but the dashboard did not use it: assets were re-sorted by category, and if you had ever tried the Custom Bucket strategy, nearly all of them were swept into the High-Risk bucket instead.
• The visible symptoms were buckets showing ₹0 assigned while one bucket held almost everything, and a High-Risk Equity+Gold figure well below what you set — because that bucket was silently holding your debt too.
• The pre-retirement de-risking glide chart then aimed at a target far lower than your buckets add up to. On one reported plan it de-risked toward 7.4% when the buckets blended to 28%.
• Because each bucket's return and volatility come from the assets inside it, this also reached the corpus projection, the Monte-Carlo simulation, the PVGA bucket sizing, the rebalance plan and the 35% equity ceiling. Expect these to move, and to move toward the plan you actually described.

Who is affected

• 5-Bucket strategy — affected. Reopen the Buckets tab and check the assigned amounts now match Edit Profile → Assets.
• Custom Bucket ladder — not affected. Those plans always read the right field; their figures are unchanged.
• 3-Bucket strategy — not affected. It never used the mechanism at fault.
• Nothing you saved was damaged. Your bucket assignments were always stored correctly; only the dashboard's reading of them was wrong.

Two ways a change could be silently lost

• Moving only the SWR slider, or switching bucket strategy without touching anything else, was never written to the vault. There was no error and no warning — the change simply was not there after a reload. Any edit alongside it saved normally, which is why it was easy to miss.
• On opening the app, the live inflation lookup could finish after your saved plan had loaded and overwrite your own return and inflation assumptions with regional defaults — which auto-save then stored as though you had made the change. A restored plan now always wins, and the lookup no longer assumes rupees for plans held in another currency.

Comfortable Retire Age could suggest an age you are not projected to reach

• If your life expectancy was set below 70, the search could return an age past it — a plan with life expectancy 65 could be told to retire at 66, on a model that assumes the money only has to last until 65. It is now capped at your life expectancy.
• The instant estimate shown before the full simulation lands also used an older refill rule that could show Bucket 1 empty while Bucket 2 was full — the same fault fixed in the projection itself in v1.1.2. Both now share one set of rules.
• 5-Bucket and custom ladder plans had this figure calculated with the 3-bucket engine. They now get an answer from the strategy they actually run.

⚠️ Some figures will read differently — they were wrong before

• The B1/B2/B3 cover figures and the Buckets tab amounts read the sliders directly. In Advanced Mode the projection derives the split from your own assets instead, so the headline could claim a 20% Bucket 1 while the simulation ran 35%. They now report what is actually being projected.
• The Corpus tab's sensitivity and delay-retirement tables projected a different pot than the headline for anyone holding gold — gold was left in the corpus and then shown again beside it. Counted once now.
• The Verdict's three what-if scenarios priced one-off spending on an old rule: an expense starting after your retirement age counted as nothing, and a genuine one-off entered as “every 1 year” was charged every year for life.
• If none of this applies to your plan, your numbers do not change.

Under the hood

• Account email, password-reset and feedback forms are now rate-limited, closing an unauthenticated path that could be used to send mail through our address.
• The app can no longer be embedded in a frame on another site.
• If you back up to a file, the pointer to it is now kept per account — on a shared computer, a second sign-in could quietly stop the first account's backups.
• A value that is not a real number can no longer be written into a saved plan as a zero.

Correction to the v1.1.3 notes

• Those notes said your browser's own Ctrl+F finds text inside collapsed help sections. It does not, and it never did — find-in-page skips hidden text. The search box in the guide does search every section, open or closed, which is what it is there for.

NPS is now modelled the way the scheme actually works

• Every asset carried a single Liquid? Yes/No switch, which can only say “spend it all now” or “never spend it”. Neither is true of a retirement account. Assets now carry a Locked till age and an Annuitised %.
• NPS Tier I defaults to age 60 with 40% annuitised — at 60, 60% comes out as a tax-free lump sum and 40% must buy an annuity. Exit before 60 and it is 80%, leaving only a fifth as cash; to model that, set the unlock age to your retirement age and the annuitised share to 80%.
• Before the unlock age the balance is not drawable — it is still yours and still compounding, it just cannot fund a year of expenses. If you retire at 52 with NPS locked to 60, the plan funds those eight years from everything else.
• The un-annuitised remainder is not lost: it arrives as a lump sum on the unlock birthday and joins your withdrawal buckets then.

The pension it buys is counted — on the grown balance, not what you put in

• The annuitised share becomes a lifetime income stream feeding the corpus projection, the Monte-Carlo runs and the Verdict.
• It is taken from the compounded balance at the unlock age. ₹10 lakh of NPS at 50 growing at 11% is ₹28.4 lakh by 60 — so 40% is ₹11.4 lakh, not ₹4 lakh, and at 6% that pays ₹68,146 a year. The rate is editable; insurers quote roughly 5.5–7% depending on the option.
• ⚠️ The payment is level for life. An Indian annuity does not rise with inflation, so it buys steadily less each year. It is also shown before tax — annuity payouts are taxable at your slab, while the NPS lump sum is not.

⚠️ This can move plans you have already seen

• If you hold NPS, your spendable corpus at retirement will be lower than before and a new pension line appears from age 60. That is the correction: the previous version let the plan spend money that cannot legally be withdrawn yet.
• Nothing changes for a plan with no NPS.

EPF is deliberately NOT locked

• This is the most common misunderstanding, so it is worth stating plainly: the EPF corpus can be withdrawn in full two months after you leave employment (75% after one), or on retirement from age 55.
• The age-58 rule belongs to EPS, the pension leg of the scheme — full pension at 58, reduced pension from 50. So EPF ships with no lock; set one only if your own situation calls for it.

Help is now searchable

• A search box over the whole guide — type “annuity”, “SWR”, “bucket” and every section that mentions it expands, including sections that are currently collapsed. (An earlier version of this note also claimed your browser’s own Ctrl+F finds that text. It does not — see v1.1.4.)
• The Advanced Mode section documents all three new fields, with a worked example.

Your cash bucket was being left empty

• On the Cash Flow table, Bucket 1 could show ₹0 for a run of years in a plan that was nowhere near running out — sometimes with tens of lakhs sitting in Bucket 2 right beside it. The money existed; the plan just never moved it across.
• The strategy allowed only one transfer between buckets per year. When Bucket 3 topped up Bucket 2, the Bucket 2 → Bucket 1 top-up was postponed to the following year. But an empty Bucket 1 means the whole of next year’s spending falls on Bucket 2, which pushes Bucket 2 back below its trigger and starts the same harvest again — so the postponement never ended.
• Bucket 1 now shows a real balance in those years. If you saw a stretch of zeros, that was the bug, not your plan.

A refill that moved nothing still used up the year’s transfer

• The Bucket 3 → Bucket 2 harvest triggers on a number of years of spending but targets a share of your corpus. When Bucket 2 sat between the two, the harvest ran, moved ₹0, and still blocked the top-up Bucket 1 was waiting for.

When Bucket 2 runs dry, Bucket 3 now tops up Bucket 1 directly

• Previously, if Bucket 2 was empty and Bucket 3 was too small to refill it, neither transfer could happen at all — Bucket 1 and Bucket 2 stayed at ₹0 for the rest of the plan while Bucket 3 quietly paid every bill.
• Bucket 3 now moves one year of spending into Bucket 1 in that situation. One year, not the full multi-year target — draining the growth bucket to fill a cash buffer is the very thing the strategy is trying to avoid.

⚠️ This can move your numbers

• Your plan now genuinely holds cash in Bucket 1 where before it left that money invested. Cash earns less than equity, so a small number of tight plans will show the corpus lasting one year less than before. In our testing that affected about 2 in 100 plans; for the great majority the final corpus is unchanged. This is the strategy working as configured, not a downgrade.

Smaller fixes

• The Buckets tab carried a verification line stating that the two refills never fire in the same year. That rule was the defect. It now confirms the thing that matters: Bucket 1 is never emptied while Bucket 2 or Bucket 3 still hold money to refill it.

Your withdrawal rate counted gold the plan had already set aside

• By default, gold is held outside the bucket strategy — the projection never spends it. But the SWR figure and the 33× Rule were divided by a corpus that still included it. The rate therefore read lower than it truly was, and the plan looked safer than it was. A plan showing 3.6% "Safe" was in fact at 4.0% "High" — the error landed exactly where the colour changes.
• The same figure feeds your Verdict score, the shortfall in Min. Corpus Needed, the suggested extra SIP and the expense-cut recommendation. All of them were flattered by the value of your gold. If you hold gold and use the default gold setting, these numbers will now read worse. That is the correction, not a change in your plan.
• Unaffected: Advanced Mode, the five-bucket and custom strategies, and anyone who routes gold into the Bond or Equity bucket — those already measured against the right corpus.

Gold you had marked illiquid was taken out of your corpus twice

• If you tagged physical gold or jewellery as illiquid, it was already excluded from your retirement corpus — and then deducted a second time. On a plan with ₹2 crore liquid and ₹30 lakh of illiquid gold, retirement was being funded from ₹1.7 crore. Your corpus will now read higher.
• Related: when gold is routed to the Bond or Equity bucket, only the gold actually inside your liquid corpus is moved there. The total was always right, but the split across buckets was not.

Comfortable Retire Age was spending gold the plan sets aside

• The same oversight reached Comfortable Retire Age, which retired you on a corpus containing gold the strategy never draws on — so it read too early. It will now generally show a later age for anyone holding gold on the default setting.

A single one-off expense no longer distorts the Min Corpus Need line

• On the Corpus chart, the Min Corpus Need line took one year's spending and assumed it repeated every year for the rest of your life. A one-off — a wedding, a car, a world trip — was therefore priced as a lifelong cost. A ₹30 lakh one-off with 25 years to run added roughly ₹7 crore to the line in that year alone.
• The line now discounts your actual year-by-year withdrawals, so a one-off is counted in the year it falls. Where your spending is smooth the line is unchanged; where you have lumpy plans it will drop sharply, and those spikes disappear. This affects the three-bucket, five-bucket and custom strategies alike.

Smaller fixes

• The Help tab quoted a fixed "33× Rule" even after you moved the Parameters → SWR slider, contradicting the dashboard beside it. It now follows whatever rate you have set. The defaults are unchanged — 3% and 33× for INR, 4% and 25× elsewhere.

Custom buckets — your own ladder

• The three-bucket and five-bucket strategies are opinions: a fixed number of buckets with researched defaults behind each one. Custom hands that decision back to you. Build a ladder of 2 to 8 buckets and give each one three numbers — how many years of retirement it funds, the return you expect from it, and how much it may swing.
• The duration is what sizes each bucket: your corpus is split by present value, so the near rungs take the largest share and the far ones the smallest. Row order is the drawdown order, so reordering the table reorders the strategy.
• The volatility you type decides how much of a bucket a crash can reach — a steady rung is largely shielded, a volatile one is fully exposed. You can override the resulting Equity+Gold weight and set each bucket's own ceiling, choose how the sizes are decided (present-value-optimal, your own percentages, or driven by the assets you tag), and set the withdrawal scheme, rebalance corridor and pre-exhaust roll-down exactly as the five-bucket strategy does.
• Nothing is blocked, but the ladder argues back: a return above about 5.5% at near-zero volatility, an implausible amount of return per unit of risk, a volatile reserve, or a volatile rung funding your earliest years each earn a warning that says which assumption a rosy projection is resting on.
• Switching between the three strategies is non-destructive — each keeps its own per-asset bucket assignments, so you can move back and forth without redoing your tagging. The Buckets, Cash Flow, Corpus, Rebalance, SoRR, Build-Up and Verdict tabs all follow whichever ladder is loaded, using your own bucket names, and the Excel template reads and writes Custom alongside the other two.

A passphrase lock for your saved plans

• Your plans have always been AES-256 encrypted in this browser under a key derived from your account. That protects them if the database is stolen off the disk — but not from anyone who can open the app as you. A passphrase closes that gap: it is mixed into the key, exists only in your head and this tab's memory, and is never stored or sent anywhere.
• You're asked for it once per visit. The key is dropped after 15 minutes of inactivity (or when you hit Lock now), and locking clears the plan from the screen as well as the key — a locked vault that still shows your numbers would be theatre.
• Setting one re-encrypts every saved plan in a single pass. There is no recovery — not by us, not by anyone. Export a JSON backup or connect a vault file first. Free with any account; Settings → Passphrase lock.

Fixes & corrections

• Corrected: the pre-retirement de-risking glide chart compared two different things. On the five-bucket strategy, your current mix and the natural-drift line are equity-only, but the dashed target line was the blended equity and gold weight — so a portfolio holding 30% equity plus 10% gold could be told it needed no de-risking when its growth exposure was really 40%. All three lines now measure equity. This moves a number you have already seen; the equity+gold figure is still shown in the Bucket De-risking Glide panel below the chart.
• The de-risking window is now yours to set. The glide chart was drawn over a fixed 10 years while the fund-movement plan beneath it ran over 7, so the two described different run-ups. One setting — anywhere from 1 to 15 years, default 7 — now drives both.
• In Advanced Mode the Edit Profile side menu could clip the last entry — the Pre-retirement SIP tab — with no way to scroll to it.

Your plans, in a folder you choose

• New Vault file in Settings: keep an encrypted copy of every saved plan as a file. Point it at a folder Google Drive, OneDrive or Dropbox already syncs and your plans travel to your other computers on their own. We never talk to Google — to the app it is simply a file, so there is nothing to authorise and no account to connect.
• Once connected it updates itself a few seconds after you stop editing. If the same file was changed on another device, the backup stops and asks you which copy to keep — it never merges two sets of plans and never overwrites the other device silently.
• Saving in place needs Chrome, Edge or another Chromium browser on a desktop. Every other browser can still download and open the same encrypted file by hand.

Corrected: Monte Carlo could score a failed plan as a success

• A plan was judged only on the corpus left at the end, so later income (a pension starting mid-retirement) could refill a corpus that had already failed to cover its expenses — and the run still counted as a success. Success now also requires that every year's spending was actually funded.
• This moves numbers you have already seen. If your plan has income that starts after you retire, its success rate may now read lower. The new figure is the correct one; re-run the simulation to see it.

Everyday improvements

• Print / PDF and an All apps switcher now sit in the top bar on every tab, and the left navigation rail can be collapsed to give charts more room.
• A save status chip beside the plan name — your plan always saved itself automatically, but nothing on screen ever said so.
• The Buckets tab now shows corridor rebalancing in the inter-bucket transfer column, so every movement between buckets is visible, not just the refills.
• Premium: the five-bucket simulation now offers the same normal and log-normal return models as the three-bucket one, and labels which model produced the result. Both start off, matching the three-bucket tab, so the two are comparable.
• Clearer wording on where tax is already accounted for.

Public launch.

Free

• Full corpus projection with a three-phase expense step-down.
• Discretionary (lumpy) expenses, each with its own recurrence, inflation and age window.
• Income, liabilities, cash flow, and the verdict on whether your money lasts.
• Monte Carlo sequence-of-returns simulation across 10,000 market paths.
• Three-bucket withdrawal strategy; Excel import and export.

Premium (Advanced Mode)

• Spreadsheet asset entry with India-specific category defaults and custom returns & volatility.
• Five-bucket, time-segmented withdrawal strategy and the Bucket Strategy Lab.
• Pre-retirement de-risking glide; per-category inflation.

⚠️ One figure changes

• Estimated tax under “Every lot you hold” charged tax on each purchase in isolation and gave losses no credit — so a position that is down overall could still show tax owed. Units redeem first-in-first-out: you choose how many, never which, so realising the winners realises the losers with them. Gains and losses now net before tax, as the realised side has always done. The estimate falls — about 2% on the book we tested, more if you hold losers.

🔬 Goal analysis

• Each goal card has a Goal analysis button. It opens the Goal Planner with that goal's current value, monthly SIP, target, asset mix, blended return and volatility already filled in — so you can try a different SIP, horizon or return without re-typing anything. Save back to Tracker returns the tuned plan into the goal form for you to confirm; nothing is saved until you do.
• The figures travel in the part of a link browsers never send to any server, and expire after an hour. No fund name, folio number or ISIN is included.

Capital Gains

• “Every lot you hold” now shows one row per position rather than one per instalment — a real book went from about 2,450 rows to about 325. Click a row for the purchases behind it and the date range they span.

Tradebooks

• Angel One, HDFC Securities, ICICI Direct and Kotak Neo tradebooks now import. Angel One and HDFC print no ISIN, so their trades come back as a filled-in template with only that column left to complete.
• ⚠️ Zerodha: please import your tradebook again. A few rows leave the ISIN cell blank and those trades were being dropped, so five holdings imported with fewer units than you bought. Blanks are now filled from the same file.
• Tickers that start with a digit (360 ONE, 3M India, 5paisa) could import as a second, separate holding.

Dashboard

• The Sync prices prompt now syncs, instead of moving you to Holdings to press a second button.

⚠️ Figures that will change

• Realised tax was overstated if you also booked a loss. A long-term loss is now set off against a long-term gain before tax, and it is the net that spends the ₹1,25,000 equity allowance — not your winning sales alone. Your estimated tax falls, and your remaining allowance rises by the loss you booked.
• Liquid and money-market ETFs (LIQUIDBEES, LIQUIDCASE and similar) were taxed as bonds. They are mutual funds: units bought on or after 1 April 2023 are short-term however long you hold them, at your slab rate. Tax on these goes up.
• Bond ETFs and target-maturity funds (“Bharat Bond ETF”, “Nifty SDL 2027 Index Fund”) were counted as equity — taxed at the equity rate, using the equity allowance, and counted as equity in your dashboard split and in every goal's drift. Your equity percentage will drop if you hold any.
• Gold funds-of-funds turn long-term after 24 months, not the 12 a listed gold ETF gets.

⚠️ ETFs showing as “Unclassified”? Import your tradebook again

• An ETF carries a fund (INF) code, and the tradebook importer used that to skip its trades, on the assumption your CAS had them. It does not — a CAS reports the folios you hold with the registrar, and an ETF lives in your demat account. The holding then had a value but no purchase dates: no return, no holding period, and left out of every capital-gains total.
• The ETF box is now ticked by default. If the same ETF also appears in a CAS you import, the CAS copy wins automatically.

Capital Gains reads by holding, not by instalment

• Realised sales and the dividend/interest ledger now show one row per holding or payer — long-term and short-term in their own columns, with the goals it funds. Click a name for every underlying sale or payout.
• Search, goal and long/short filters on the realised table, the income ledger and the open-lot table. A new panel shows how the year's tax comes out of the year's gain, bucket by bucket.

Since v1.2.1

• Dashboard: the three source cards open into drill-downs — funds by fund house, demat by broker, everything else by platform.
• Discover: AMFI stopped printing each fund's plan and option. Direct · Growth funds went from about 1,800 to over 3,000, and 2,745 income-distribution plans that were being served as growth plans are now separated.

⚠️ If your mutual funds stopped updating, or read as worth nothing

• AMFI added two columns to the daily file every Indian mutual fund in this app is priced from. The app was reading the price from where it used to be, so it stopped recognising more than half of the schemes AMFI lists, and priced most of the rest at zero.
• Nothing failed visibly and nothing was lost — the units and transactions in your vault were never touched. Only the prices were wrong, which is why it looked like a portfolio that had quietly collapsed.
• Please run a NAV sync (Import → Sync prices). It repairs every affected holding in one pass. If a fund still reads zero afterwards, it is a genuinely stale scheme, and the app will now say so instead of showing a zero.

Fund research works again

• The same AMFI change moved each fund's plan and option out of its name, so no scheme said "Direct" any more — and Discover found nothing at all, over a list of more than fourteen thousand funds that was entirely present.
• Fixed, and it now reaches 1,750 Direct · Growth funds across every fund house and category. The app also alerts us automatically the next time AMFI changes this file's layout, rather than waiting for someone to notice.

⚠️ The Risk tab was measuring less than half your portfolio

• It reported honestly on what it covered — but it covered about 57% of holdings and under half the value, and blamed the gap on missing price history. That was not the reason.
• When several statements described the same fund, a blank identity from one could overwrite the real one from another, and a fund you had sold was never repaired afterwards. Both are fixed, and the app now restores the identity of exited holdings on the next sync.
• On a real portfolio, coverage went from about 54% to 86% of holdings. Your risk figures will change, because they are now measured across far more of your book — volatility, drawdown, beta and capture will all read differently.

⚠️ EPF: every year now shows, and the return was understated

• Importing several years of passbooks only ever listed the current financial year's contributions. All imported years are now kept and shown together, with a running total and the financial year each one falls in.
• EPF credits interest once a year, not monthly. The return was being solved against today's balance, which on any date before the credit lands is a balance that has not been paid its interest yet — so the figure read roughly a third too low. It is now measured to the last point the account was actually settled.
• The holdings table and the pop-up were also answering the question two different ways and disagreeing. There is now one rule, used everywhere.

Bonds: a repayment is an exit

• Wint Wealth's master report imports directly now.
• Most bonds never have a sell row — they repay their principal and that is what closes the position. The app now retires units against principal repaid, so matured bonds stop showing as holdings you still own.
• Coupons still due this financial year are shown as a forecast, clearly separated and kept out of your taxable income totals — they have not been paid yet.

Open a fund's details from anywhere

• Click a fund's name on any tab — dashboard, holdings, SIPs, transactions, buckets, risk, rebalance or capital gains — to open its transactions and performance. The pop-up now also names the folio or broker it belongs to, so two holdings of the same fund are told apart.

Smaller things

• Capital Gains now names the holdings it cannot compute, instead of leaving a silent gap — with a link straight to the transaction it needs.
• Please request your CAS with zero-balance folios, since inception. Without them the statement leaves out funds you have fully sold entirely, and no app can report on what it was never given. The Import screen now says so.
• The Risk tab explains its own arithmetic — a worked example that runs the real calculation on sample data, so you can check what Sharpe, Sortino and beta actually mean here.
• Capture ratios drew the wrong bar. A holding that gains while its index falls could show a full-length bar, or the bar belonging to the goal you looked at before. The percentages beside them were always right.
• Adding the same share twice no longer creates two rows that never add up.
• A broker file could produce phantom folios from its summary pages. Fixed.

⚠️ A stock split made your return jump

• The price feed has already restated a share's whole history for any split or bonus — a pre-bonus ₹2,708 close is republished as ₹1,354. The app was also multiplying your share count by the same factor, so one event was corrected twice.
• A 2-for-1 therefore printed as a +100% day on the performance chart, and every value before the split showed as half what the holding was actually worth. Mutual funds were never affected — a fund's published NAV is never restated, so there the adjustment was already correct.
• Your numbers will change: a share that has split or issued bonus shares while you held it will show a lower, correct return, and a flat line where the jump used to be.

Dividends are found and recorded for you

• The exchange record carries a share's dividend history, and the app was already asking for it and discarding it. Payouts against shares you demonstrably held are now added to your income ledger automatically and flagged for you to check.
• Two things need your eye, and the app says so rather than guessing: the dates are ex-dates, not the day you were paid — and the year you were paid is the year that counts; and TDS is left at zero, because the ₹5,000 threshold is per company per year.
• Dividends are not capital gains. Since FY2020-21 they are taxed as income at your slab rate, so they are kept out of the capital-gains figures — the Capital Gains tab now says where they are instead.
• A holding whose trades don't add up to the quantity you hold claims nothing at all: a partial tradebook cannot say what you held on a 2019 ex-date, and a plausible wrong tax figure is worse than none.

Corporate actions, all in one place

• One screen for the whole portfolio: every split and bonus recorded against your shares while you held them, each labelled with what was done and why — applied, suggested, already in your book, or nothing to do.
• Where two events could explain the same discrepancy — Reliance split 2-for-1 in both 2017 and 2024 — the app asks rather than guesses, because which one it was changes your tax lots.
• Everything is reversible: Undo sits beside Confirm, at the same size.

Per-holding performance

• Click any holding's name to open its transactions, its performance and its corporate actions. Subscribers see what their own position did — value over time with a marker on every purchase and sale, the rupee gain, and its share of the portfolio's total gain.
• Free for everyone: the instrument's own price history over 1D to Max.
• ⚠️ The chart draws what the position is worth; the percentages beside it divide out your own deposits, so adding money never shows as a gain.

US holdings, in rupees

• Vested and INDmoney positions are now measured using a daily USD/INR history, so the currency half of the return is included — a stock down 8% in dollars can be up for you.

Smaller things

• Invested value beside current value on the dashboard, holdings and every drill-down.
• Family drill-down — the whole family, or one member at a time.
• Adding a transaction to a hand-entered holding now updates its quantity. Previously the units stayed put, so buying more didn't show.
• Stock search by name against the NSE/BSE list, filling in the ISIN and today's price.
• Backup reminders if auto-backup isn't set up, and the auto-backup folder is now shown in Settings, not just the file name.
• Upcoming SIPs no longer lists instalments whose dates have passed.

Columns you can size, headers that stay put

• The holdings table sized its columns to whatever was in them, and no cell was allowed to wrap — so one long fund name pushed Current, Gain and XIRR off the screen behind a horizontal scrollbar. The numbers you came to read were the ones that left.
• Every column now keeps the width it is given, a long name wraps instead of shoving its neighbours, and you can drag any column edge to resize it. Your widths are remembered in that browser; “Reset column widths” restores the defaults.
• Heading rows now freeze while you scroll, on Holdings, Transactions, Rebalance, Buckets and Capital Gains.

Sovereign Gold Bond coupons are now recorded

• An SGB pays 2.5% a year on its issue price, in cash, every six months. That money never appears in the bond's value — which is why the value shown is correct — but it is real return the app recorded nowhere, so SGBs looked like a pure gold tracker.
• “Record SGB coupons” on the Capital Gains tab rebuilds the whole payment history from the tranche calendar. It counts only coupons paid after you bought (an SGB bought on the exchange did not earn the ones paid before you owned it), at the units you held on each payment date.
• It is safe to run again — it replaces its own previous entries and never touches one you typed. A bond with no purchase date, or whose imported trades don't add up to the units held, is reported rather than guessed at.

⚠️ Dividends and coupons now count toward your return

• Recorded income joins the XIRR calculation on the dashboard, goals, buckets, family and rebalance. If you have any income recorded, your reported returns will go up. They were understated before — a payout that reached your bank was simply missing from the return it was part of.
• Only dated receipts count. The cumulative “interest so far” figure a bond statement carries is excluded, because putting years of interest on one date would produce a nonsense number.

Rolling returns: a chart that runs forwards

• Comparing two funds of different ages produced an x-axis reading 2025, 2026, 2016, 2017… — the younger fund's dates laid down first and the older fund's appended after, so the two clouds sat side by side instead of overlaid on the dates they share.
• It is now a real time axis: one shared, proportional scale, every point at its true date.

A more accurate portfolio return

• The dashboard's portfolio XIRR is now one calculation over your actual cash flows — every purchase, every redemption and every date, solved once. It previously combined each holding's own return into an average, which does not describe a portfolio: individual returns are annualised over the different lengths of time each holding has been held, so they are not directly comparable with one another.
• This is the same method already used for each individual holding, and for goals and buckets, so the dashboard figure is now consistent with the rest of the app.
• Your portfolio's headline return may therefore read differently from before. Nothing about your investments has changed — only the way the single summary figure is worked out.
• Every other return in the app — per holding, per goal, per bucket — was already computed this way and is unchanged.

Discover funds — put several funds together and measure the mix

• Fund groups, free: give a group a name, put funds in it with a weight each, and see what that mix would have done — its return, how much it swung, its worst fall, and its rolling returns over every start date. Four fund cards side by side cannot answer this: returns average out, volatility does not.
• Choose how often it is rebalanced — never, quarterly, half-yearly or yearly. These are genuinely different portfolios with different returns, so it is a control rather than an assumption, and the screen says how many resets the interval took. Each one is a real sale, with an exit load and a tax bill that the figures do not count.
• Where your weights ended up. A mix set at 50/30/20 and left alone for ten years is not a 50/30/20 portfolio at the end — the table shows both.
• A monthly SIP amount backtests what those instalments would have bought: money in, what it would be worth, and the return per year. History, never a forecast.
• How alike have these funds behaved? A grid comparing every pair. This is not stock overlap — it says how alike two funds have moved, which is a different question and the one that can be answered honestly from published prices. It also appears when you compare individual funds.
• Starting shapes — two-fund core, core and satellite, three-cap split, equity and debt, India and abroad. Each sets only the slots and the proportions: no preset names a fund, and you choose every one yourself.
• Groups are saved in your encrypted vault and are included in your JSON backup and your vault file.

Fixes

• Upside and downside capture used to vanish silently when they could not be measured. They now say why: capture needs at least twelve months in each direction, counted separately, so a one-year period can never show it and a three-year one often shows only one side.
• The period and rebalancing controls are now labelled on screen, rather than being two unexplained rows of buttons.
• The sample portfolio contained an investment dated 1926, which blanked the portfolio return for anyone browsing without an account.

Discover Mutual Funds — a new tab, free

• Search any open scheme AMFI lists, by name, fund house or SEBI category.
• See what it did: returns, rolling returns over every start date, volatility, Sharpe, Sortino, worst fall, beta, alpha and upside/downside capture.
• Compare up to five funds on one chart, with Direct or Regular named per row.
• Research, not recommendation. Nothing is ranked, scored or suggested.

Risk tab

• Pick your benchmark — Nifty 50, Next 50, Midcap 150, Smallcap 250 or Nifty 500 — instead of one fixed index.
• Alpha and upside/downside capture added, plus a plain-English summary of how the portfolio compared with the index.
• The benchmark now appears on the ₹100 chart and on the rolling-returns table, and buckets get a ₹100 chart of their own.
• Beta and alpha per holding in the side-by-side view.

Fixes

• Alpha compared the index over the full window while the portfolio was measured from its first purchase — the difference was reported as alpha.
• The growth and drawdown charts covered different periods, and both drew a flat stretch before the first holding existed.
• A benchmark that predates its tracking index fund is now refused, not measured over the years it does cover.
• The sample portfolio had scheme codes belonging to other funds, so its demo figures came from the wrong NAV history.
• A new release now announces itself instead of waiting for a hard refresh.

Your XIRR says what you earned. Nothing in the app said what that return cost you along the way. Two portfolios both returning 13% are not the same proposition if one of them halved twice getting there, and until now there was no way to tell them apart.

A new Risk tab (Advanced)

• Six figures, measured on your own daily portfolio value — reconstructed from your transactions and each holding's daily NAV, not copied from a factsheet. Volatility (how bumpy the ride was), Return, Sharpe (reward per unit of bumpiness), Sortino (the same, counting only the falls), Worst fall — the deepest peak-to-trough drop you actually sat through, with its dates and whether you have recovered — and Beta, how much of your movement is simply the market's.
• The number no factsheet can give you. A portfolio's volatility is not the average of its funds' volatilities: holdings that fall on different days partly cancel, so the real figure is lower. That gap is what diversification is actually worth to you, and it only exists once someone reconstructs what your book was worth on every single day — which needs the transaction history sitting in your own vault.
• Rolling returns. Your headline return depends on the day you happened to start. Each row slides the same holding period across every trading day in the window and reports the whole range: the worst stretch you could have picked, the average across every start date, the best, how often it made money at all, and how often it beat the risk-free rate. Read the worst column first — it is the one that tells you whether you could have sat through this portfolio.
• Two charts. What ₹100 would have done, with your contributions divided out; and an underwater plot showing how far below its own best-ever value the portfolio was on each day. (That second one never rises above zero by design — zero means "at a new all-time high", not "no gain".)
• Windows of 1, 3, 5, 7 or 10 years.

Four ways to slice it

• Your whole portfolio — one honest figure for everything you hold.
• Each holding, side by side — one risk profile per fund, ETF or share, grouped by kind so a liquid fund is never ranked against a small-cap share, and one row per fund rather than per folio number. Pick up to five and chart them against the Nifty, with quick picks by category so you can line your large caps up against each other, then your small caps. These deliberately do not add up to your portfolio's figure — comparing a row against it is how you see what your diversification is worth.
• Every bucket — each group measured on its own daily series. There is no total row: volatilities do not add up, and a bucket that looks calm on its own may not be.
• Any goal — the money assigned to one goal, at the share assigned to it.

What it will not pretend to know

• It says what it could not measure. EPF, PPF, fixed deposits and property have no daily price and never will, so they are left out of the figure entirely rather than counted as flat — counting them as flat would make your portfolio look calmer than it is. The panel names every holding it left out and why, and reports the share of your value it could actually measure.
• Beta is withheld when it would not mean anything. A portfolio that is mostly debt or gold has a beta against an equity index that is arithmetically defined and completely meaningless. Below a minimum fit, the number is replaced by the reason.
• These are time-weighted figures and will not match your XIRR, on purpose. Money you paid in is divided out, so they describe how your holdings behaved rather than when you happened to buy — which is what makes them comparable with a fund factsheet. Your XIRR is money-weighted and answers the other question. No number already in the app has changed.
• All of it is backward-looking. Every figure describes what already happened; none of them predicts anything.

Also

• A new FAQ section explains each measure in plain terms — what Sharpe and Sortino actually are, what counts as good, why the drawdown chart never goes above zero, how to read rolling returns, and what to do when the numbers look bad (mostly: nothing).
• Price history is downloaded once, then cached in your vault. The first run fetches each holding's daily NAV with a progress count; after that the tab opens instantly and only tops up the last few days. Nothing about your portfolio leaves the browser — only anonymous scheme codes are requested.

Sovereign Gold Bonds

• Your SGBs now price themselves. One bond is one gram of 999 gold, and that is exactly how they are now valued — against the daily IBJA 999 rate, the same benchmark the RBI redeems them at. This works whether the bond came from a demat import or you typed it in by hand; for a hand-entered one, the number of grams is all that is needed.
• Fixed: a Sovereign Gold Bond could show a price years out of date. SGBs trade rarely, so a market data feed often returns the last price anyone paid — which for one bond was from January 2019, around 73% below what the gold inside it was worth. Worse, it was stamped with today's date, so nothing looked wrong. Prices that are more than a month old are now rejected outright rather than shown as current. If you hold SGBs, expect their value to jump when you next refresh prices — that is this correction, not a market move.
• Tell your bond apart by its tranche. Enter a symbol like SGBJUN31I and the app identifies which of the 67 issues it is, along with its interest payment dates and maturity. It is optional — leave it blank and the bond is still valued correctly.

Property, REITs and InvITs

• ⚠️ REITs and InvITs are no longer counted as equity. Listed business trusts — Embassy REIT, India Grid Trust and the like — arrive from a demat statement with nothing in their name to identify them, so they had been filed as ordinary shares, putting office parks and toll roads in your equity total. They now have their own Real estate category, identified from the security type recorded in the ISIN. This applies to holdings you imported long ago, not just new ones, so your asset allocation may look different the next time you open the app without your having bought or sold anything. Listed NCDs are corrected the same way, from equity to debt.
• Capital gains on REIT and InvIT units are treated as Section 112A assets — twelve months, 12.5%, and the ₹1.25 lakh yearly exemption shared with your shares and equity funds, not a second allowance of their own.
• Record a property's loan and its rent. Outstanding balance, EMI and final instalment date, plus monthly rent and running costs — and the app works out the growth rate since you bought, the gross and net rental yield on what it is worth today, and how long the loan still has to run.
• Net worth after debt, shown alongside your portfolio. Your portfolio total stays what you own — every allocation, goal and rebalance figure is measured against it, and netting a home loan off would quietly distort all of them — so the two are reported side by side rather than merged.

Dashboard

• Everything you own is on it now. EPF, PPF, fixed deposits, property, gold bonds and anything else entered by hand were missing entirely, so the opening screen showed your mutual funds and demat holdings and called that your portfolio.
• Your goals at a glance — how many are on track and which need attention — and the SIPs due next, summarised by category and by goal. The instalment-by- instalment list stays on the Active SIPs tab, which is where you would go to act on one.
• Portfolio health now shows you where the marks are. Rather than a bare score, it names the two measures with the most points available — what each one reads, and which direction moves it — and a new FAQ section explains exactly how the score is worked out and why it can change on a day you didn't trade.
• Switching tabs takes you to the top of the new one instead of leaving you part-way down it.

Buckets

• A new Buckets tab. Group your holdings the way you actually think about them — Core / Growth / Experimental, or Equity / Debt / Real estate — and give each group the return you expect of it. Every bucket then shows its own realised XIRR, its gain, its asset mix, and whether it is ahead, on target or behind what you signed it up for. A goal tells you what money is for; a bucket tells you what job it is doing.
• A holding can be in more than one bucket. A flexi-cap fund is both “Equity” and “Core”, and both readings are useful — so buckets are lenses, not folders. Each bucket's own figures are exact; where the same money sits in two of them, the screen says how much, so a column that adds up to more than your portfolio never reads as an error.
• Sleeves inside a single goal. Separately from the portfolio-wide set, each goal can be split into buckets of its own — “the equity sleeve of Retirement”, “the property sleeve” — so you can see which part of a goal earned its return rather than one number for the whole thing. These belong to that goal alone, and the same fund can be grouped differently in another. Shown on the goal card and on the Rebalance screen.
• Group holdings from wherever you happen to be: tick rows in Holdings and use 🪣 Set bucket, open a bucket and tick what belongs in it, or work down one goal's holdings one at a time.
• Property, EPF and imported snapshots are handled honestly. They have no transaction history to solve a true XIRR from, so they report an estimated return — always labelled “est.”, and never quietly averaged into a real one. Where an EPF or NPS statement already carries a solved return, that is the figure used.

Returns

• A holding you have sold out of now reports the return you actually made. Closed positions are solved from their own buys and sells. A part-imported trade history that is missing its opening buys still shows “—” rather than a number built on half the story.
• Bonus issues and stock splits. Record one and your units, cost basis and return all stay correct across it — no cash changes hands, so nothing is treated as a purchase. Your holding period is preserved for capital-gains purposes, and the app can spot one for you when an imported quantity jumps but the cost doesn't.
• Holdings less than a year old now show a plain, un-annualised return (“abs · 7 mo”) beneath the gain, so a few strong months aren't read as a yearly rate.

Fixes & additions

• Sovereign Gold Bonds have their own type in the add-holding form. They imported correctly all along; entering one by hand meant choosing between “Bond” (which filed a gold-linked asset under Debt) and “Physical gold” (which has no maturity date). Thanks to the reader who pointed this out.
• Fixed: the Feedback form could silently drop your message. It said “Sent”, but the spam trap behind it was being auto-filled by browsers and password managers on real submissions, and those were discarded. If you wrote to us and never heard back, that is why — please do send it again.
• Fixed: renaming a bucket or a family member now reaches your vault file. Backups were scheduled by changes to holdings, goals and transactions only, so an edit that touched nothing but a name or an expected return didn't trigger one, and could be missing from a restore on another device.
• Fixed: “Refresh prices once a day automatically” switched itself off the moment you switched it on. Ticking it started two things at once — the setting itself, and, because the setting turning on is what the daily refresh waits for, that refresh's own record of having run. They were saved to the same place at the same moment and the second overwrote the first, so the box stayed ticked until you next opened the app and prices never refreshed on their own. Please turn it back on; it works now. Restoring a backup no longer resets it either. Thanks to the reader who reported it.

• Fixed a same-day buy and sell throwing off a holding's units and return. A stock bought and sold on the same day could leave phantom units behind, inflating the quantity, value and XIRR. Corrected everywhere; affected holdings fix themselves the next time the app loads.
• The “Edit transactions” return now matches the Holdings table for demat stocks. When your imported trades don't yet cover a stock's full quantity, both hold the return at “—” until the history adds up, instead of one showing a figure the other doesn't.
• A reminder when a holding's statement is over a month old, so a recent SIP instalment or purchase isn't quietly missing — with the last import date shown per holding.
• Optional once-a-day price refresh. Turn it on in Settings and prices sync in the background when you open the app.
• The monthly SIP flowing into each goal, on the Goal cards and the Rebalance screen.

US stocks & ETFs

• Import your Vested and INDmoney statements. Your US order book becomes per-stock transactions, with each trade converted to rupees at the USD/INR rate on the day it was placed — so the cost and the return carry the currency move as well as the price move. These are treated as International holdings (24-month long-term capital gains), and “Sync market value” keeps pricing them in dollars.
• One file, both markets. The tradebook template now has a separate “US Trades” sheet. Fill the Indian sheet, the US sheet, or both, and everything imports in one pass.
• US and foreign holdings by hand. The Stock / ETF form takes a currency — pick USD and enter the dollar price and cost, with the exchange rate now and on the day you bought it, and it stores the rupee figures with a full audit trail.
• International equity (including RSUs) and unlisted / pre-IPO shares are now added under Stock / ETF rather than as a separate “other” investment.

Returns you can trust

• The return on a holding now matches the one in its transactions. After a price refresh, the XIRR in the Holdings table could lag the price it was solved against — most visibly on US holdings, where the price and the exchange rate both move. Returns are now re-solved against the latest price on every sync, so the table and the “Edit transactions” preview always agree. Your existing figures correct themselves the next time the app loads.
• Gain and XIRR never disagree on direction. A US holding down in dollars can still be up in rupees once the rupee has weakened; the gain and the return now always tell the same story, because both are costed from the rupees you actually paid, trade by trade.
• A realised return per goal. Each goal shows its money-weighted XIRR — what the bucket has actually earned over its cash flows — beside the blended forward estimate, on both the Goals cards and the Rebalance screen.

Fixing what couldn’t be priced

• Edit straight from the “couldn’t be priced” list. Every row now has an Edit button, and the list stays open so you can fix each unpriced holding in one sitting rather than reopening it after every one.

Corrected: an auto-contributing balance grew every time you saved it

• Opening an EPF, PPF or NPS account that contributes automatically, changing nothing, and pressing Save added another run of contributions on top of the projected figure. Doing it four times took a ₹7.9 lakh balance to ₹23 lakh. The form now edits the balance you last saw on your statement, which is what its own hint always said it was.
• If this happened to you, the balance will drop back to its correct value. Nothing was lost — the inflated figure was never real. Older records that are missing an "as of" date are re-anchored to today the next time you save them, so check the date on any such account once.

New

• Owner on the manual holding form. Adding a holding by hand always filed it under the account holder, whoever the Family filter was showing — there was no way to say whose it was without correcting it afterwards. Now you choose, on both adding and editing.
• When each price was last fetched. A "Prices as of…" line on the Dashboard and the Holdings toolbar, a per-holding tooltip with the exact time, and an amber dot on anything not priced for over four days. Only for holdings we actually fetch a price for — an EPF balance or a flat is a figure you typed, not a quote.
• All apps menu in the header: the other InvestApps tools listed directly, one click away, on desktop and phone.

Family

• Removing someone now asks first, and says what happens to their money: their holdings — and the transaction history behind them — go back to the account holder, so nothing is deleted and no return or capital gain is lost. You can send them to a different person instead if you'd rather.
• The Edit and Remove links no longer sit flush against each other.

New retirement-account and tradebook imports, and a merge that gives the same result whatever order you import in.

New imports

• NPS: import your CRA transaction statement — one line under Other investments, expandable into its equity / corporate-debt / government-debt schemes, with a cash-flow-solved XIRR.
• EPF: import your EPFO member passbook (PDF) — the Employee / Employer / Pension breakdown and a cumulative XIRR that stays correct even when a partial year is imported on top of a full one.
• Broker tradebook (buy/sell history): your trades become transactions with a real XIRR and FIFO capital gains. A partial tradebook that doesn't reach your first purchase reads “—” until its trades reconcile with your holdings, rather than a made-up return.
• Bond holding statements from Wint Wealth and GoldenPi, with their real cost basis and the interest paid.

Imports that agree with each other

• Any order: importing your CAS, broker holdings and tradebook now produces the same XIRR whatever order you do them in.
• A fund that appears in both your CAS and a broker statement is counted once — the CAS copy, which carries the transactions, wins.
• The most recent broker holdings statement always sets today's quantity, even if an older one is imported after it.
• Duplicate detection: re-importing the same holdings or tradebook is recognised and flagged before anything is saved, so a position is never doubled.

Public launch.

Free

• Dashboard: mutual funds, stocks, bonds, EPF, PPF and deposits on one screen, with the true equity/debt split.
• Import: CAMS and KFintech CAS, broker holdings and bond statements, parsed inside your browser — nothing is uploaded.
• Holdings with a real, cash-flow-solved XIRR you can edit; Active SIPs.
• An encrypted vault with backup to a folder you choose, and a one-click delete.

Premium

• Goals: map holdings to a target and get the SIP that closes the gap.
• Rebalance against the glide path.
• Capital Gains: FIFO long-term vs short-term, a tax estimate, and what's left of your ₹1.25 lakh long-term exemption.
• Family: who owns what across the household, and where the equity risk sits.