It is the fifth of the month in a research room in Mumbai. The SEBI PMS monthly report sheets are finally merged, the merged cells are unpicked, and the product head wants one slide by lunch: who grew last month, and why. The AUM column is sorted, a manager is circled in green, and nobody has yet checked whether that growth came from clients or from the market.
Our view: SEBI PMS monthly report data is most useful when you read flows before AUM, and least honest when a share figure hides what it was divided by. If your team already runs a PMS data platform that collects every portfolio manager’s monthly filing, the analysis below is mostly arithmetic. If you are still collecting by hand, start with our guide to what the SEBI Portfolio Manager Monthly Report contains, then come back here.
AUM growth tells you less than flows do
A manager’s AUM can rise because clients brought money in, or because the portfolios went up in value. Those are different stories. One says the business is winning clients. The other says the market was kind this month.
The monthly report gives you the pieces to separate them. It carries AUM by service, inflows and outflows, clients by category, turnover, transactions, schemes and complaints. Most slides use one of those numbers. We would use at least three.
Four numbers answer most questions
For each manager, each service and each month, keep four fields side by side: opening AUM, closing AUM, inflows and outflows. If the filing does not give opening AUM directly, take it from last month’s closing figure, which is why keeping history matters. From those four you can derive everything a product or distribution head usually asks for.
| Derived measure | How to calculate it | What it answers |
|---|---|---|
| Net flow | Inflows − outflows | Did clients add money or take it out? |
| AUM change | Closing AUM − opening AUM | How much bigger or smaller is the book? |
| Other change | AUM change − net flow | How much of the change was not client money, mostly portfolio value moving |
| Net flow rate | Net flow ÷ opening AUM × 100 | Is the flow large relative to the manager’s size? |
| Share of total | Manager AUM ÷ total AUM of managers in scope × 100 | How big is this manager against the set you collected? |
Do these per service. Discretionary, non-discretionary and advisory behave differently, and a single blended total hides which one is moving. Decide up front whether advisory assets belong in your totals, and write that decision on the sheet. Also check the unit the report states before you add anything up.
A worked example: two managers, one month
The numbers here are illustrative and rounded, not market data. Say two managers each report discretionary AUM growing from 500 to 540 crore in a month. On an AUM-sorted slide they look identical.
Manager A had inflows of 60 crore and outflows of 35 crore. Net flow is 60 − 35 = 25 crore. AUM change is 540 − 500 = 40 crore. Other change is 40 − 25 = 15 crore. Net flow rate is 25 ÷ 500 × 100 = 5%.
Manager B had inflows of 20 crore and outflows of 22 crore. Net flow is 20 − 22 = −2 crore. AUM change is still 40 crore, so other change is 40 − (−2) = 42 crore. Net flow rate is −2 ÷ 500 × 100 = −0.4%.
Same closing AUM, opposite stories. Manager A is winning client money. Manager B is losing a little of it, and the market is covering the gap. A slide that circles both in green has answered the wrong question.
Two managers can end the month at the same AUM and be telling you opposite things about their clients.
A share figure is only as honest as its denominator
“Manager A holds 2% of the industry” sounds precise. It is only true if the total underneath covers the industry. When collection is manual, it rarely does. Teams track the twenty or thirty managers they have time for, and the share is quietly a share of that list. We wrote about why that matters in tracking only some managers skews the picture.
Two habits keep share figures honest. First, put the denominator on the slide: “share of the 30 managers collected” or “share of all managers with a filing this month”. Second, keep managers who filed nothing out of both the numerator and the denominator, and mark them as “no data filed” rather than zero. A missing filing summed as zero shrinks the total and inflates everyone else’s share.
Clients and complaints need a per-client view
Raw client counts favour large managers, and raw complaint counts punish them. Neither tells you much on its own. Divide instead. Average AUM per client, by client category, shows who serves which kind of investor. Complaints received per 1,000 clients puts a small manager and a large one on the same scale.
Turnover and transactions deserve the same care. Read them against AUM, not as absolute numbers, and compare a manager with its own history before comparing it with peers. A sudden jump against its own last six months is often more informative than a ranking.
The case for just using industry aggregates
The fair objection is that industry totals are already reported in the press and by industry bodies. Why rebuild them from each manager’s filing, eight sub-reports at a time?
For a headline, aggregates are fine, and we would use them. They stop being enough when the question gets specific: which managers drew net inflows in discretionary last quarter, how flows split by client category, or whether a peer’s growth was flows or market. Aggregates cannot be segmented that way, and they cannot be traced back to a manager when a number looks wrong. Rebuilding from filings is more work. It is also the only way to answer the questions a product or distribution team actually asks.
Build the dataset once, then analyse every month
The analysis itself is simple arithmetic. What makes it hard is the dataset underneath: every manager, every month, numbers stored as numbers, managers matched by registration number rather than by name, and each month kept as a dated version so last month’s closing AUM is still there when you need it as this month’s opening figure.
That is the work Terminal X takes off the desk. Its SEBI PMR agent collects every registered portfolio manager each month, marks empty filings as “no data”, reconciles scheme counts and keeps source, fetch time and version on every record. Once flows are clean, you can join them to strategy performance using our PMS returns vs benchmark dataset guide.
What to do this week
- Add opening AUM, inflows and outflows next to closing AUM in your current monthly sheet, per service.
- Compute net flow and other change for every manager, and re-sort by net flow rate instead of AUM.
- Write the denominator on every share figure, and count how many managers it covers.
- Mark managers with no filing as “no data filed” and exclude them from totals.
- Pick one peer and check whether last quarter’s growth was flows or market.
None of this changes the source data. It changes which questions the data can answer honestly. More on building research datasets is in the Terminal X guides.
This post is about research data workflows, not investment advice. For any investment decision, confirm with your adviser.