Most people don’t exactly think “fun” when they hear the words finance management. However, most people tend to believe that keeping track of personal expenses helps them stick to a budget and whom am I to argue? Finance apps must have something going to them.
This article is going to focus on the technology used in the finance mobile app development and the finance app market as a whole. Hopefully, this will help us get some insight in this area.
We are going to look at two types of finance apps — simple and complicated. In simple apps users enter all their purchases manually, so I’ll use the same term as CBN — “manual entry” apps. Conversely, “complicated” personal finance applications link to a user’s bank accounts and thus I’ll be referring to them as linked apps. So how to develop finance apps?
Manual Entry Finance Apps
It’s no secret that the average mobile user is a lazy animal and prefers everything wrapped and ready. However, the nature of “manual” apps requires users to do all the work by themselves. That’s why such applications need more than anything to be simple and user-friendly. Otherwise, they won’t be popular.
Math is all “manual entry” apps do, but not all that personal finance app developers like Yalantis can do. Such apps normally track expenses by putting them into categories and then reporting them. One great example is Spendee. It has really a nice graphic interface and doesn’t overwhelm users with categories, letting them choose among a number of options, such as food & drinks, entertainment, car, shopping, and others. The app is protected by a user-selected password that must be entered every time the app is used. But security in the “manual entry” application is not the biggest concern for the developer, since a user’s bank accounts aren’t linked to the app.
Manual entry personal finance app can make users’ life easier by letting them take pictures of their receipts. Receipt scanning is accessible through Optical Character Recognition (OCR) technology, which helps to turn receipts into data. When this is done, the system converts the data taken from the receipt into useful graphs that show users how they’re spending their money.
Lemon Wallet, which was acquired by LifeLock, offered receipt digitalization technology in their app. They used a machine learning system to identify information on the receipt, an OCR system to extract the needed data (an amount, a date, a merchant name, etc.), and a human aid and supervision system to edit the entry.
Expensify automatically scans data from receipts and enters it in a user expense record. The app is a variation on the mass finance app market. It offers establishments a great tool for reporting employees’ expenses during business trips and targets business users looking to get reimbursed. The receipt you scanned is saved in the app, where it will then be converted into a digital copy.
You can check the list of the OCR libraries here. However, recognition quality is not perfect on mobile devices and the text quality of the average receipt is rather low. What is more, OCR libraries tend to be large, which means that a mobile application will be of considerable size if you use OCR for offline recognition. The mobile platform cannot provide the same performance as desktop, therefore using an OCR engine on a mobile device is unusual. Apps that apply OCR technology usually communicate with cloud-based APIs.
Linked Finance Apps
You are probably aware that Mint Personal Finance dominates the market of budget and finance management apps. It has been cited by Wired, CNET, Digital Trends, The New York Times, and everybody else as the best finance mobile solution. I wrote an article that uncovers its underlying technology which you can read here. Developing this kind of app requires a lot of attention to security, since it implies collecting and ferrying data about user income and transaction history from financial institutions to your finance app, which will then analyze user spending automatically. If you are serious guys promising bank-level security to your customers, keep reading to learn how to create an app like Mint.
Financial kings like Mint employ established players, such as Intuit (Mint’s owner) and Yodlee, to do the account plumbing. These systems compile all sorts of accounts, such as bank, investment, credit card, loans, and others into a single place by means of account aggregation practice.
All user sensitive finance data gets stored in the system. You can define spending allotments with corresponding spending alerts and limitations, visualize checks and transactions, and provide finance modeling and strategizing through cross-account trend analysis and foresight.
Yodlee-powered finance sites typically do not store credentials, even when the data goes through the site’s servers. They only pass them through to Yodlee, which stores the data securely. Things are a bit different when it comes to Mint, which applies Intuit’s proprietary system and stores the data on its own servers.
Typically, a user’s finance data is pulled from his or her bank accounts using a data-stream format called Open Financial Exchange (OFX). It was developed by Microsoft, Intuit, and CheckFree back in 1997 and supports a wide range of finance activities including consumer and small business banking, bill payment, investments, including stocks, bonds and mutual funds, and tax download.
OFX enables a direct connection between the client and a financial institution’s server through request / response model using XML. Financial institutions normally have APIs that transmit OFX data so that users can access it, but not all of them do. In this case you can actually build your own API and pull the data into your service, just like Yodlee and Intuit did: if you go down this path though, I can only imagine how complicated your life will become. Any company can potentially build their APIs according to OFX’s specifications. It’s free and open source. We have never tried doing that, but from what I’ve seen on the web it’s not the best idea. OFX is old and a mess.
You can also try Open Bank Project. It claims to provide an open source API for banks that you can use to build applications and services based on the account holder’s transaction data. You can see its architecture on Github.
OR you can integrate with one of the existing systems.
Available Account Aggregation Systems
Yodlee is the leader when it comes to account aggregation systems. It covers finance accounts not only in the US, but also in the UK, France, Spain, South Africa, Australia, New Zealand and India. However, it’s expensive and startups generally can’t afford it. In 2009 TechCrunch published an article talking about Mint’s relationships with Yodlee, where they stated that in those days total payments from Mint to Yodlee were around $2 million/year. So you can get the feeling of their prices.
Intuit is cheaper, than Yodlee, and costs about $0.30 Per User Per Month. Other options include CashEdge (which is now a part of Fiserv), Finicity, and ByAllAccounts, a web bank account aggregation app which is more fitting for finance advisors. I also found a Spain-based platform for European market called Eurobits.
Read also: Local delivery apps market
Since finance and accounting apps are mostly free and the price for the account aggregation services is high, some startups are taking extraordinary measures to survive. I came across a highly disturbing piece on The Guardian, which features Britain’s most popular personal finance app OnTrees and a similar app called Money Dashboard. Both apps run on the US-based platform Yodlee. In exchange for their free services, you are effectively granting them access to your personal finance data, which they use to monetize. In other words, they make money by selling insights into user spending habits.
Some apps don’t like this monetization strategy and prefer not making money at all. Budgeting app Level says they do not monetize by selling user data or through advertising. So they are still figuring out ways to gain revenue. By the way, the Level app looks awesome!
Read also: Quality standards in Yalantis
Other finance planning apps like Mint monetize through lead generation. They collaborate with different establishments and advertise their services within the app by offering users ways to save money. This is a very cool way to make money, I’d say, but also a heavyweight feature.
The Bgdt “manual entry” budget app from Sweden is paid. It costs $1,99. Spendee, as I mentioned before, relies on an advertising model and in-app purchases to make ads go away on my Android. Interestingly enough it costs $1,99 on Apple App Store.
The calendar-based personal finance app Dollarbird has a Pro version available via in-app purchase for $4.99 per month, or $49.99 per year. This feature is temporarily unavailable though.
To summarize, there’s a lot of room for creativity in finance application development, both in terms of features and monetization.