Digitale Banken: Die Zukunft des Privatkundengeschäfts und der Vermögensverwaltung

Interaktive Portfolio-Übersicht mit Tableau
Interaktive Zusammenstellung eines Portfolios mit automatischer Gewichtung auf Grundlage der Sharpe-Ratio

Vor gut einem Jahr habe ich mit dem Blog-Post „Digitale Banken: Welche Anforderungen bringt die Digitalisierung?“ einige Ideen aufgezeigt, wie sich Banken die Digitalisierung zu Nutze machen können. Die Motivation ist meist die Steigerung des Umsatzes, Kostenersparnis und die Erschließung neuer Märkte. Die Herausforderungen sind nach wie vor schwindenden Zinserträge, härtere regulatorische Vorschriften, erhöhter Wettbewerb und anspruchsvolle Kunden.

Seither war ich an vielen weiteren Projekten im Finanzwesen beteiligt und sehe, dass besonders das Privatkundengeschäfts und die Vermögensverwaltung von der Digitalisierung profitieren. Dabei habe ich drei Kernbereiche ausgemacht, die ich nachfolgend etwas mehr in den Fokus rücken möchte.

1. Payment und Zahlungsverkehr

Vielen Menschen fällt es schwer sich die sperrige IBAN zu merken. Die Email-Adressen, die Handynummer und den Fingerabdruck haben die meisten allerdings immer parat. Neue technische Möglichkeiten treffen so auf veränderte Verbraucherbedürfnisse. Zahlungen werden immer häufiger mit dem Smartphone abgewickelt – ohne Eingabe einer IBAN. Zahlungen werden schneller – kein ganzer Tag vergeht zwischen Kontobelastung und Gutschrift, sondern nur wenige Sekunden.

Außerdem werden Zahlungen dezentral. Dazu befasst sich der Finanzsektor zunehmend mit der Blockchain-Technologie (wie hier die Deutsche Bank), mit welcher nicht nur Überweisungen sondern auch Wertpapierhandel möglich sind. Insbesondere große Institute tun sich mit der Blockchain-Technologie allerdings schwer, da die zentrale Kontrolle des Zahlungsverkehrs auch als Instrument zur Kundenbindung verstanden wird.

2. Finanzierung und Kreditvergabe

Bei der Finanzierung und Kreditvergabe stehen neue/bessere Algorithmen im Mittelpunkt. Mit diesen möchte man stets folgende Fragen beantworten: Wie lässt sich der Credit Score optimal ermitteln? Wie empfehlen wir unseren Kunden proaktiv das richtige Finanzprodukt?

Die Konstruktion von Entscheidungsbäumen, eine Form des Maschinellen Lernens, ist für solche Algorithmen eine tragende Säule. Hierbei ist es ratsam die Ergebnisse immer wieder zu kontrollieren und gegebenfalls über Parameter die Gewichtung bestimmter Variablen zu korrigieren. Sonst kann es womöglich passieren, dass einem Gutverdiener der Dispo gestrichen wird, weil sein Gehaltseingang nicht als solcher erkannt wird; oder einem Rentner ein Bausparvertrag angepriesen wird, weil sein Social-Media-Profil einen solchen Bedarf vermuten lässt.

3. Beratungsplattformen für die Vermögensanlage

Zu Zeiten der Null-Zins-Politik ist bei der Vermögensanlage zunehmend Kreativität gefragt. Diese lässt sich schwer in starren Systemen aus dem Zeitalter der Mainframes abbilden. Stattdessen lassen sich Anlagestrategien aus unterschiedlichen Blickwinkeln mit Analytics-Anwendungen, wie Tableau („Datenanalyse für Banken„), beleuchten und sich ggf. um quantitative Funktionalität aus R anreichern.

Beispielsweise lässt sich mit Hilfe der Sharpe-Ratio interaktiv eine optimale Portfolio-Gewichtung berechnen, und die Williams-Percent-Range als Indikator für Chart-Trends nutzen. Ein solcher Self-Service-Gedanke passt zur Unternehmenskultur der Datendemokratisierung und lässt auch die Mitarbeiter aus Vertrieb und Beratung an der Digitalisierung teilhaben.

 

Alle hier gezeigten Beispiele sind echten Szenarien nachempfunden. Die Portfolio-Optimierung kann als Tableau Packaged Workbook (twbx) hier heruntergelanden werden, und benötigt Tableau mit R-Integration. Auch hierzu freue ich mich wieder über Feedback, Fragen und Anregungen…

KPMG Global Automotive Executive Survey 2016

KPMG Global Automotive Executive Survey 2016: click to open interactive story
KPMG Global Automotive Executive Survey 2016: click to open interactive story

In the recent months, 800 automotive executives from 38 countries gave their insights to KPMG. You can discover the key highlights of the KPMG Global Automotive Executive Survey in this eye-catching interactive Tableau story.

This is a fabulous example how you can use stories to present a narrative to an audience. Just as dashboards provide spatial arrangements of analysis that work together, stories present sequential arrangements of analysis that create a narrative flow for your audience.

How to use a custom Mapbox map as your background map in Tableau

Mapbox map in Tableau
Mapbox map in Tableau

Tableau now comes with more geographical data built in, including updated US congressional districts (CD), local name synonyms for world capitals, Japanese postal, and Mapbox integration. I’ve to admit I really love Mapbox!

What is Mapbox? Mapbox is an online repository of custom-built maps for your needs and enables you to create the perfect map to integrate into your Tableau visualization. Mapbox maps are highly customizable – you can design your own map, build applications, extend applications, use satellite imagery and create static maps. You can even have Pirate Maps!

Mapbox tutorial:

  1. So first off you’ll have to register with mapbox.com
  2. Once you are logged in go to Account > API access tokens > copy and paste your token. You’ll need this for Tableau.
  3. Open up Tableau > connect to your data source that has geographical locations. For this case, we will use the sample sales data set that is preloaded in Tableau.
  4. Map > Background Maps > Map Service to open a popup box.
  5. Add > Mapbox Services > Classic
  6. Fill in a style name for this map > Paste in the access token you previously copied
  7. Drop the selection box down and it will provide a list of classic maps already for your use. For this case, we are going to use Emerald.
  8. Take your city dimension from the data set > double click or drag and drop to populate a map. See below the before and after without Mapbox and with.

If you create multiple Mapbox maps and want to populate different styles on different worksheets, you can:

  1. Maps > Background Maps > Emerald. Here you have a list of maps that you have created.

So here you have a basic understanding of using Mapbox in Tableau.

Happy mapping, literally go explore! And join me on Twitter:

How to speed up Tableau by using Performance Recordings

Tableau Performance Recording Timeline
Tableau Performance Recording Timeline

Getting your dashboards up to speed can be quite difficult if you don’t know where the latency is situated. The first and most important rule about making workbooks more efficient is to understand that if it loads slowly in Desktop on your computer, then it will be slow on the server too once it is published. Tableau Desktop and Tableau Server each have their own way to enable, record, and analyze performance.

A must have for performance tuning your workbooks. All you have to do is start the Tableau Performance Recording, make your workbook action and stop the Performance Recording. A few seconds later, Tableau opens a new workbook with the Performance Summary dashboard in it.

Create a performance recording in Tableau Desktop

  1. To start recording performance, follow this step: Help > Settings and Performance > Start Performance Recording
  2. Make some dashboard operations and/or refresh your data source(s).
  3. To stop recording, and then view a temporary workbook containing results from the recording session, follow this step: Help > Settings and Performance > Stop Performance Recording
  4. You can now view the Performance Summary dashboard and begin your analysis.

Create a performance recording on Tableau Server

  1. Administrators must enable the feature. This is located under settings, for each site.
  2. Check the box and save for Workbook Performance Metrics.
  3. Navigate to a view on the server.
  4. Remove the iid=xx from the URL.
  5. Enter in its place record_performance=yes. Your full URL should now look something like this: https://data.alexloth.com/#/site/AA/views/Superstore/Summary?:record_performance=yes
  6. After the page reloads, you’ll notice the ID is added automatically back to the URL and that a performance button appears within the View’s toolbar. Don’t click on the performance button yet.
  7. Do some filtering and some clicking within the workbook such as applying filters, selecting marks/rows, and clicks that cause actions to other elements of the visualization.
  8. Then click the performance button.
  9. Now you’re ready to click on the Performance button which will launch a new window with the Performance Summary dashboard.
  10. Don’t forget to disable the performance recording in the admin settings when you are finished.

Understand the Performance Summery dashboard

The Performance Summery dashboard contains three views:

  • Timeline: a Gantt chart displaying event start time and duration.
  • Events sorted by time: a bar chart showing event duration by type.
  • Query text: It optionally appears when clicking-on an executing query event in the bar chart.

Time line Gantt chart

The uppermost view in a performance recording dashboard shows the events that occurred during the recording, arranged chronologically from left to right. The bottom axis shows elapsed time since Tableau started, in seconds.

In the Timeline view, the WorkbookDashboard, and Worksheet columns identify the context for the events. The Event column identifies the nature of the event, and the final column show each event’s duration and how it compares chronologically to other recorded events.

The events sorted by time

This section of the workbook shows the duration of recorded events in descending order. This is useful for observing the execution time of each event that occurs during the performance recording. This will help you identify any lengthy events that may be the cause of performance problems.
Events with longer durations can help you identify where to look first if you want to speed up your workbook.

Different colors indicate different types of events. The range of events that can be recorded is:

  • Computing layouts: If layouts are taking too long, consider simplifying your workbook.
  • Connecting to a data source: Slow connections could be due to network issues or issues with the database server.
  • Executing query: If queries are taking too long, consult your database server’s documentation.
  • Generating extract: To speed up extract generation, consider only importing some data from the original data source. For example, you can filter on specific data fields, or create a sample based on a specified number of rows or percentage of the data.
  • Geocoding: To speed up geocoding performance, try using less data or filtering out data.
  • Blending data: To speed up data blending, try using less data or filtering out data.
  • Server rendering: You can speed up server, rendering by running additional VizQL Server processes on additional machines.

Query text

Alternatively, the workbook also displays the query text for any specific event that you want to examine in detail. You can access the detail by clicking on any of the green executing query events in the bar chart. This is a handy feature which allows you to review any query text that may be of interest without having to leave the tableau performance summary dashboard.

If you click on an Executing Query event in either the Timeline or Events section of a performance recording dashboard, the text for that query is displayed in the Query section.

Data Science Toolbox: How to use R with Tableau

Recently, Tableau released an exciting feature that enhances the capabilities of data analytics: R integration via RServe. By bringing together Tableau and R, data scientists and analysts can now enjoy a more comprehensive and powerful data science toolbox. Whether you’re an experienced data scientist or just starting your journey in data analytics, this tutorial will guide you through the process of integrating R with Tableau.

Step by Step: Integrating R in Tableau

1. Install and start R and RServe

You can download base R from r-project.org. Next, invoke R from the terminal to install and run the RServe package:

> install.packages("Rserve")
> library(Rserve)
> Rserve()

To ensure RServe is running, you can try Telnet to connect to it:

Telnet

Protip: If you prefer an IDE for R, I can highly recommend you to install RStudio.

2. Connecting Tableau to RServe

Now let’s open Tableau and set up the connection:

Tableau 10 Help menu
Tableau 10 External Service Connection

3. Adding R code to a Calculated Field

You can invoke R scripts in Tableau’s Calculated Fields, such as k-means clustering controlled by an interactive parameter slider:

SCRIPT_INT('
kmeans(data.frame(.arg1,.arg2,.arg3),' + STR([Cluster Amount]) + ')$cluster;
',
SUM([Sales]), SUM([Profit]), SUM([Quantity]))
Calculated Field in Tableau 10

4. Use Calculated Field in Tableau

You can now use your R calculation as an alternate Calculated Field in your Tableau worksheet:

Tableau 10 showing k-means clustering

Feel free to download the Tableau Packaged Workbook (twbx) here.

Connect and Stay Updated

Stay on top of the latest in data science and analytics by following me on Twitter and LinkedIn. I frequently share tips, tricks, and insights into the world of data analytics, machine learning, and beyond. Join the conversation, and let’s explore the possibilities together!

Blog post updates: