# Mapping the market and portfolio data stack
At 4:00 PM New York time, the closing price for a single share of Apple flashes across a Bloomberg terminal. Within minutes, that one number ripples through a chain of systems inside an asset management firm: it updates the value of thousands of client portfolios, feeds a fund's official price, and lands in reports that regulators, auditors, and investors will trust as fact.
That journey, from a raw price feed to a trusted portfolio record, is the heart of the asset and wealth management (AWM) data stack. Let's trace it.
Market data is the raw pricing and reference information about securities: prices, yields, volumes, corporate actions, and identifiers. It comes from vendors and exchanges.
The big names you will hear:
A firm rarely relies on a single source. A common practice is to designate a primary source and one or more fallback sources. If Bloomberg does not deliver a price for a thinly traded bond, the system reaches for the next source in a defined hierarchy.
Why this matters: market data is expensive. Vendor contracts are among the larger line items in an operations budget, and usage is tightly licensed. Displaying a Bloomberg price on a client-facing screen may require a different license than using it internally.
Keep these straight, because the whole stack depends on the distinction:
Reference data lives in the security master.
The security master is the central database that defines every instrument a firm can hold or trade. Think of it as the firm's official dictionary of securities.
Each entry answers: what exactly is this thing? For our Apple share, the record stores:
The same security carries different codes in different systems. One vendor calls it "AAPL US Equity," another uses the ISIN, a third uses an internal ID. If these do not mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.Voir la définition complète → cleanly, a price update can attach to the wrong instrument, or to none at all.
This is why security master data management is a discipline of its own. A break in identifier mapping is a common root cause of pricing errors.
A useful free reference on global identifier standards is the Association of National Numbering Agencies, which oversees the ISIN standard.
Once we know what a security is (security master) and what it is worth (market data), we need to know how much of it we hold. That is the job of the position-keeping system, often part of a broader portfolio accounting or investment book of record (IBOR) platform.
An IBOR is the firm's real-time record of positions: every holding, across every portfolio, updated as trades settle and cash moves.
For our Apple position, the position keeper tracks:
When a portfolio manager buys Apple, the order flows through an order management system (OMS), executes, and then settles. Settlement is when cash and shares actually change hands. In most major equity markets this now happens one business day after the trade (known as T+1), following the 2024 move to T+1 in the US, Canada, and Mexico.
The position keeper reflects the holding, and the security master identifies it, and the market data values it. Three systems, one position.
Now the payoff. The NAV (net asset value) is the per-share value of a fund. For a mutual fund or ETF, it is calculated at least once a day and is the price at which many investors buy and sell.
The formula is simple:
NAV per share = (Total assets - Total liabilities) / Shares outstandingThe complexity is in the inputs. Striking the NAV means:
1. Pull the closing price for every holding from the agreed primary source.
2. Match each price to the correct security via the security master.
3. Multiply price by quantity from the position keeper.
4. Sum all holdings, add cash, subtract fees and other liabilities.
5. Divide by shares outstanding.
For our fund, Apple's closing price times the number of shares held becomes one line in that sum.
NAV is not struck once and trusted blindly. Firms run a price validation or tolerance check process. If today's price for a security moved more than a set threshold (say, a large percentage swing) versus yesterday, the system flags it for a human to review. Was it a real market move, or a bad data point?
Many funds also use a fund administrator, an outside firm that independently calculates NAV as a check against the manager's own number. When the two match within tolerance, confidence is high. This independent oversight is a core investor protection.
Vérification des acquis
1. What is the fundamental distinction between market data and reference data in the AWM data stack?
2. Why do asset management firms typically designate a primary source and one or more fallback sources for market data?
3. The excerpt describes a closing price rippling through 'thousands of client portfolios,' a fund's official price, and regulatory reports. What core concept does this illustrate?
4. Select ALL correct answers about why market data licensing and cost matter in operations.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers that correctly classify examples of reference data (static) versus market data (dynamic).
Sélectionnez toutes les réponses correctes.
Understanding failure points is what separates fluency from theory.
Stale prices. A market holiday abroad, or a halted stock, means yesterday's price carries over. Good systems tag prices with a timestamp and source, so a stale value is visible, not hidden.
Corporate action errors. A missed stock split is a classic disaster. If the price divides by four but the share count does not multiply by four, the position value drops by 75 percent overnight. Corporate action processing is heavily automated and heavily monitored.
Identifier mismatches. As noted, a broken mapmapUsing software to automate repetitive marketing tasks and campaigns, enabling personalisation at scale across channels like email, web, and social.Voir la définition complète → between an ISIN and an internal code can orphan a price update. Reconciliation catches these.
Reconciliation is the process of comparing two independent records and investigating any differences. AWM firms reconcile constantly:
A break is any unexplained difference. Operations teams spend much of their day chasing breaks to zero before NAV is finalized. This is unglamorous and absolutely central.
Regulators increasingly expect firms to prove not just that a number is right, but that they can explain where it came from. This is data lineagedata lineageData lineage maps how data moves and transforms across systems, from origin to consumption, showing where it came from, what changed it, and where it goes.Voir la définition complète →: the documented path of a data point from source to report.
For our Apple price, lineage would show: Bloomberg feed, timestamp, security master match on ISIN, applied to the position keeper quantity, included in the NAV calculation, published in the fund fact sheet.
Strong data governancedata governanceData governance is the set of policies, roles, and processes that ensure data is accurate, secure, well-defined, and used responsibly across an organization.Voir la définition complète → (the policies, roles, and controls over data) turns a tangle of systems into an auditable chain. It is also what lets a firm adopt AI and automation safely, because you cannot trust an algorithm if you cannot trust its inputs.