User statistics

More than seven hundred users, one hundred fifty financial institutions, downloaded the FIBO Data Model. This article analyzes downloads by data modeling tool, institution type, and geography. The impressive size of the user community influences the data model and by extension the Enterprise Data Management Council (EDMC).

Daily downloads

The orange line of totals is surprisingly linear, climbing at around two new users per day.

FIB-DM daily downloads (column chart)
FIB-DM daily downloads

The columns show daily downloads including zero activity on weekends and other days. Spikes often happen when there are FIB-DM news or publications, like the newsletter on 22 January 2020.

While there is no poll data, feedback suggests that data architects find the FIBO model from Google Search, LinkedIn, mouth to mouth, and now via the EDM Council website. The Financial Industry Business Data Model is already outpacing competing vendor industry models, and there is no plateau in sight.

I have been a speaker at FIBO conferences and in discussion with EDMC for many years. However, the council’s decision to feature FIB-DM as an external FIBO Product on their website is thanks to you, the users.

Financial Institution Type

FIB-DM Institutional downloads (pie chart)
FIB-DM Institutional downloads

One hundred and six different financial institutions and their regulators downloaded the FIBO Data Model.

Financial Markets are the major FIBO content area with ontology modules for Securities, Derivatives, and Indices in production, and Corporate Actions, Collective Investment Vehicles (Funds), and Market Data in development.

Surprisingly, Investment Management/Servicing, Brokerage, and Financial Markets constitute only 17% of institutional downloaders. Three quarters are Banks and Credit Card Companies. Pointing out the large banking user base, I am asking to prioritize and accelerate banking content like Loans, in discussions with the EDMC. However, this requires more active participation from banks in the FIBO content groups.

For diversity, I want to see even more Regulators, Government Agencies, Multilateral Financial Institutions (development banks), and non-profits.

I didn’t do a breakdown by institution size yet, as this would require to look up balance sheet information of the banks. However, my impression is that most downloads are from midsize banks.

Data Modeling Tools

FIB-DM Data Modelling Tools (pie chart)
FIB-DM Data Modelling Tools

PowerDesigner, ERWin, and Sparx EA are the data modeling tools most widely used with FIB-DM. The largest segments, PowerDesigner and ERWin are also the most widely used tools in the financial industry.

The FIB-DM Masterfile is a PowerDesigner Conceptual Data Model. Hence 40% of users use that tool. However, most other data modeling tools, including ERWin, can import native PowerDesigner model files.

When the FIBO Data Model released in fall last year, some disappointed Sparx Enterprise Architect and IBM InfoSphere Data Architect users told me that their tools cannot import PowerDesigner model. The statistics made me realize that Sparx, IBM IDA, comprise 20% of modelers interested in the industry standard. I created FIB-UM, the universal model, a UML-XMI download that works. This is a prime example of user statistics driving product development.

 

Geography

Users from all continents (except Antarctica) downloaded the model.

FIB-DM Downloads by Region (pie chart)
FIB-DM Downloads by Region

The chart breaks down 210 FIBO Data Model users by region. European banks made the most downloads. The “Global” slice comes from multinational banks and the large global consulting companies.

I like the diversity and global interest in the FIBO data model. There should be even more users from South America and Africa!

Here is a list of FIB-DM countries: Angola 1, Argentina 1, Australia 17, Austria 1, Belarus 1, Belgium 9, Bermuda 1, Brazil 11, Bulgaria 1, Canada 31, Chile 1, China 6, Columbia 2, Cyprus 1, Czech Republic 2, Denmark 6, Dominican Republic 1, Dubai 1, Egypt 1, El Salvador 1, Emirates 5, Estonia 1, Finland 1, France 6, Germany 11, Ghana 1, Global 61, Greece 1, Honduras 1, Hong Kong 3, Hungary 4, India 17, Indonesia 2, Iran 1, Ireland 1, Italy 1, Japan 8, Jordan 5, Kenia 1, Korea 1, Luxembourg 2, Madagascar 2, Malta 2, Mexico 7, Morocco 1, Nepal 1, Netherlands 13, New Zealand 5, Nigeria 1, Norway 4, Oman 1, Peru 1, Poland 2, Portugal 2, Russia 9, Saudi Arabia 1, Singapore 10, Slovakia 2, South Aftrica 15, Spain 6, Sweden 11, Switzerland 8, Thailand 1, Turkey 8, UK 33, Ukraine 1, Unknown, USA 93, Venezuela 2, Vietnam 8

East Asia is still unrepresented. Just a few users from China and Japan and still no downloads from Korea and Taiwan!

Data Source and quality

The download form log is the underlying data for statistics.

Download form

When FIBUM, the UML-XMI version came out a user on the Sparx EA forum asked: Why do I have to provide my email in order to download the model?

  1. The secure download link sent via email limits the proliferation of the model file, including defective or modified versions across the web. (Bots can’t reach the encrypted model file).
  2. I send updated download links and the newsletter to the email list.
  3. Freemium to Premium, I individually contact banks that catch my attention. Sure commercial licensees are funding the Open Source model. I want institutional users to upgrade to the Full Model.
  4. The domain part of the email address identifies the company.

Email address

The domain part of your office email, right of the ampersand, is the URL that uniquely identifies your company. The full email address also is the primary key for the download log. In other words, I can remove duplicates from multiple form submissions.

Alas, almost a third of downloaders use a generic (e.g. Gmail) address. Many users want to check out the model at home first or must do so because their firewall prevents file downloads. In these cases, the form’s company field helps to classify institution type and country. However, there remain 22% of all submissions with unknown company type or country.

Non-institutional downloads, mostly consulting companies constitute 37% of downloads. They range from individual contractors, small and medium-sized companies to large global IT and Business Consultancies.

There can be double-counting when a user submits the form twice, with her office and personal email.

Other modeling tools

There are eighty submission with the data modeling tool stated as “other”; half of these have a blank moment field. I would think that many where downloads by mistake, but so far only one email requested “remove” from the newsletter.

In summary, only two-thirds of submission records are high quality data to aggregate region, modelling tool and institution type. However, there is no indication that the charts would change substantially, if submission data was complete.

Downloads vs. users

To the best of my knowledge, there are around Two hundred fifty active downloading users. That is the total number of form submissions minus remaining duplicates and one-off downloads.

Some very large banks and consultancies had several employees downloading the model. I usually connect them, so they can leverage migration and other work already done.

From meetings with banks, I learned that many downloaders put FIB-DM on ERWin ModelMart, the PowerDesigner or another repository, or an intranet fileserver. It is not unusual for me to find half a dozen data modelers at a bank attending the FIB-DM overview meeting.

More users may have received the data model file from a colleague or associate as an email attachment.

I estimate that one active downloader creates on average 2-3 active users.

That puts the total of active FIB-DM modelers between 500 and 750.

Appeal for data quality

Please

  1. Use your office email in the download form or to subscribe to the newsletter.
  2. Long-term contractors, or personal emails, put the FI you work at in the company field.
  3. Take care to specify your “other data modeling tool.”
  4. Use the newsletter subscription form to update your info!

Conclusion

More than seven hundred users, one hundred fifty financial institutions downloaded the FIBO Data Model. There are between 500 and 1,200 data modelers actively using FIB-DM.

The impressive size of the user community influences the data model and, by extension, the Enterprise Data Management Council (EDMC).

The community is global with users in Europe, Asia, Afrika, and the Americas.

Thanks for reading. I appreciate your feedback.

Jurgen Ziemer, jziemer@jayzed.com