This latest chapter of our Artificial Intelligence-saga in the Asset Management value chain is dedicated to its impact on client services. Robo-advisors or mobile app are only the tip of the iceberg in terms of Asset Management industry’s digital mutation. And even in client relationships, AI allows to give answer to customers’ attempts such as immediate answering or bespoke products.
To read previous episodes, click here
Chatbots & Virtual agents: investor’s new assistants
They are called Leo at Generali, Erica at Bank of America, Djingo at Orange Bank, HARO at Hang Seng… Virtual assistants are experiencing an accelerated development in banking and insurance sectors. According to Deloitte, companies pursue four main objectives in their implementation: improve customer acquisition ; increase revenue per user ; reduce churn by increasing customer satisfaction and offer a 24/7 support ; minimize cost to serve. A Juniper Research report has found that the adoption of chatbots across the retail, banking and healthcare sectors will realize business cost savings of $0.70 per interaction, for an annual total of $11 billion by 2023, up from an estimated $6 billion in 2018.
Business application in finance is more recent, most likely because the field is more complex than answering a question about account situation. As Forbes wrote in 2014 about Kensho’s algorithm – recently sold to S&P Global and which scan million market data points to find investment opportunities – “it can find answers to more than 65 million question combinations in an instant by scanning more than 90,000 actions.” [1]
And the learning time is as greater as the data is complex. Thus, it took Bank of America nearly 18 months and 100 dedicated people to enable Erica to acquire its learning capabilities (through Natural Language Processing and Machine Learning) and to be able to interact through the voice, chat or tap (guided conversation via buttons) with the million customers seduced within 2 months after its launch.
Only a few Asset managers have launched their interface...
Because of this technical complexity, most of asset managers’ initiatives are currently in test phase. However, some firms stand out from the crowd.
Schroder Investment Management officially released in June 2018 the full version of its online chatbot, Schroders GO, in Singapore after completing a testing period and updating its features with four Singapore-based intermediaries, including Citibank and UOB. The company claims it is the first asset manager in Singapore to launch such a chatbot facility to its distribution clients.
Schroders GO, which operates through Facebook Messenger, offers real-time communication to the firm’s intermediary client base, allowing them to access information on all mutual funds recognized for sale in Singapore, including non-Schroders funds, as well as financial markets and market views, without the need to download an additional app.
The chatbot was already able to answer different enquires, such as “What is the net asset value of the Schroder Asian Income?”, “What did the FTSE close at?” or “What is the performance of the BlackRock Asian Dragon Fund?”. Its full version has added features, including investment-linked policies data and information on the Central Provident Fund Investment Scheme. Schroders has also further optimized the presentation of fund performance data, which include non-Schroders funds, charts and market information.
Schroders will gradually make its chatbot available to the company’s other intermediary clients in Singapore, to be followed by an expansion to other markets in the region.
DWS has also started investigating how to use AI to improve sales techniques and attract clients. The asset manager has exchanged data sets with several technology companies specialized in AI analysis. The company wants to use AI to introduce chatbots that can respond to simple questions from retail clients when they come to its website.
In its 2017 Business review, BNP Paribas Asset Management announced that it's testing a chatbot to guide users in their fund search. This tool is able to provide product details, compare them and explain technical terms. At the end of this test phase, it will be able to assist the advisors of their networks, professionals and end customers. In its latest financial presentation, the Group indicated that 17 chatbots have been deployed notably in Wealth Management (myChat&Trade, myVirtualAssistant, etc.). To read our article on BNP Wealth, click here.
… while some providers take a position on B2B2C market
Because of investment engagements required to develop chatbots, some providers propose ready-to-implement platforms, where virtual agents are pre-trained with industry and domain knowledge.
IBM Watson – one of the leaders in AI industry – has launched a chatbot starter kit for companies in the investment and wealth management sector. The offering helps companies get started on the path to creating a virtual assistants that enable users to query portfolios and associated holdings and also offers a simulated analytics service for securities performance under a given scenario.
We can also cite Dorsum, an Hungarian software company, leader in Central and Eastern Europe and expanding globally. Its platform for customer onboarding have recently been awarded in the “Best Chatbots Platform” category in Juniper Research’s Future Digital Awards.
The Fintech has also developed a wealth management mobile app, which integrates the chatbot DAISY (for Dorsum Artificial Intelligence System). The product grants an hybrid advisory model, building a bridge between artificial intelligence and personal advise. The client is able to talk to a self-learning chatbot or send a request to a financial advisor to take over the conversation.
Experiments have started on speaking assistants
Even if texting is a widespread way to communicate, especially for Millennials, voice is becoming increasingly common to interact with smart devices. More than 56 million smart speakers will be sold in 2018 [2] and two-third of smartphone owners have already tried its voice assistant. The next step of the chatbot’s development will certainly be the “Voicebot”.
Since March 2018, JPMorgan’s clients are able to ask Amazon’s Alexa for the bank's research reports. And the firm is also working on financial assets price data requests. On JPMorgan Asset Management side, it is now possible to get the latest market insights only asking “Open Market Insights” using Alexa or Google Home. In December, SBI MF launched similar service in India with Alexa where all your mutual fund related queries can be addressed easily.
More experimental, UBS created a digital clone of its chief economist. The project, named UBS Companion, is aiming to explore how to create a new access to UBS's expertise for clients and to test the use of human digital assistants in a wealth management context.
100 client and 10 advisors in Wealth Management Switzerland have started the experiment with two newly developed avatars: "Daniel Kalt", a digital version of UBS Regional CIO Switzerland and “Fin”, a friendly helper and digital assistant. The avatars, which can interact via voice and eye contact with clients, are being deployed to client meetings via a TV screen. Tapping IBM's Watson technology, the fake Kalt uses information given to it by the real Kalt to answer clients' more complex questions about the market.
Reporting: automated writing capabilities are speeding process away
Digital platform are increasingly providing clients – retail, institutional, intermediaries and advisors – with tools that make it possible to realize on-demand, real-time, self-service and bespoke financial reports. And more interactivity with CRM allow salesforce to integrate to reports images, video or links to marketing material. The acquisition of 100M, a digital reporting solution Fintech, by NeoXam, a leading provider of data management and transaction software solutions dedicated to the financial industry, is a recent witness of this evolution.
But, fund commentaries redaction is often on the critical path of the production of real-time financial reports. New AI-powered solution – named Natural Language Generation – appeared in the early 2010s. Narrative Science Inc., launched in 2010 with computer-generated news articles, added products for financial-services businesses in 2013 ; those firms now account for 60% of the company’s client base. Other startups offering automated reports for financial-services firms include Automated Insights, Yseop, Capital Cube, ARRIA (integrated to IBM cloud) or S&P backed Kensho Technologies Inc. Recently, these solutions have started to be implemented by the Asset Management industry.
Yseop, Capital Cube, ARRIA (integrated to IBM cloud) or Kensho Technologies Inc. Recently, these solutions have started to be implemented by the Asset Management industry.
Societe Generale Securities Services (SGSS) and Addventa announced in October a partnership to launch a service for automated drafting of portfolio management commentaries based on artificial intelligence solutions. Management companies, who are SGSS clients, will benefit from an automated and instant drafting of performance commentaries for their financial investment portfolios which cover a given time period, selected by the client, in different languages, in a clear and consistent style. Addventa also partnered with Allianz GI and BNP Paribas AM, the latter stating that 90 portfolio commentaries are automatically generated each month.
Nine of the top 25 asset managers use Narrative Science’s Quill. Franklin Templeton selected it to enhance efficiency of fund commentary process. American Century Investments is incorporating Quill’s ability to write up rudimentary fund recaps, while T. Rowe Price is working on using the AI-powered software in some of the firm’s investment-strategy descriptions. A unit of Fidelity Investments is exploring similar technologies to see if they can turn data into personalized communications for consumers.
As Franklin Templeton’s Vice President says, automating portfolio commentaries enable companies to scale delivery of more routine fund content and data updates across multiple strategies and markets, while allowing them to dedicate more expert resources to authoring in-depth investment perspectives and thought leadership material.
Author: Pascal Buisson - November 2018
[1] https://www.forbes.com/sites/antoinegara/2018/03/06/wall-street-tech-spree-with-kensho-acquisition-sp-global-makes-largest-a-i-deal-in-history/ - [2] Source: Canalys
