# A common data format

DevFeed: [A common data format](<https://devfeed.tech/articles/a-common-data-format-30538.md>)

Original publisher: [Read original article](<https://medium.com/build-addepar/a-common-data-format-e3e4b684e47d?source=rss----596e43e5e150---4>)

Author: Dénes Findrik

Published: 2017-10-17T13:11:10Z

Content type: article

Language: en

Sources: [Addepar](<https://devfeed.tech/sources/addepar.md>)

Topics: [data](<https://devfeed.tech/topics/data.md>), [Finance](<https://devfeed.tech/topics/finance.md>), [Graphs](<https://devfeed.tech/topics/graphs.md>), [CSV](<https://devfeed.tech/topics/csv.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>)

Tags: [csv](<https://devfeed.tech/tags/csv.md>), [data](<https://devfeed.tech/tags/data.md>), [etl](<https://devfeed.tech/tags/etl.md>), [finance](<https://devfeed.tech/tags/finance.md>), [fintech](<https://devfeed.tech/tags/fintech.md>), [text](<https://devfeed.tech/tags/text.md>), [transactions](<https://devfeed.tech/tags/transactions.md>)

## AI overview

The article explains how financial data arrives from many sources in inconsistent formats and proposes a simple common interchange format. It argues that separating shared accounting concepts from data-transformation processes can help standardize inputs to a Financial Graph and make the platform easier to build on and validate.

## Source excerpt

Every day we parse financial data from hundreds of sources, including custodial banks, prime brokerages, fund administrators, private banks, and market data vendors. We receive data in a variety of formats: CSVs, TSVs, text files, custom formats -- and there are basically as many representations as sources. This variety of data formats is a side effect of Finance's long history with Tech: the industry has been a major force of innovation. Financial institutions have invested a lot to ensure that they have the technology they need to operate. As a result, competing complex systems are now used to represent the same data. Each different system introduces different constraints, and each constraint can have implications for how people think about the inner workings of the financial world. How can we push all this data into a consistent Financial Graph? The answer lies in defining a simple common format. The Problem Receiving data in a lot of different formats isn't necessarily a hard problem on its own: the greater challenge is handling the significant differences in how the same data is represented from source to source. To better illustrate this problem, let's take a very simple example. We have a single account that already has 100 shares of Alphabet and $15,000 cash. From that account, we buy 20 shares of Apple at $140 per share with a $50 broker fee for the transaction. The following are just a few hypothetical ways this same, simple set of transactions could be represented at the end of the day: Source 1 Positions.csv Date ,Account,Symbol,Name ,Quantity,CurrCode ,Type,Exchange 20170131,1234 ,GOOG ,Google Inc,100 ,USD ,STK ,NASDAQ 20170131,1234 ,USD ,US Dollars,12150 ,USD ,CRY , 20170131,1234 ,AAPL ,Apple Inc ,20 ,USD ,STK ,NASDAQ Transactions.csv Date ,Account,Quantity,Amount,Symbol,CurrencyCode,Type,SettleDate 20170131,1234 ,20 ,2800 ,AAPL ,USD ,BY ,20170131 20170131,1234 ,0 ,50 ,AAPL ,USD ,FEE ,20170131Source 2 Positions.txt Account Date Symbol Quantity 1234 2017