5 strategies for delivering better financial experiences.

There is significant opportunity in the financial industry to improve the financial experiences we provide to our customers. A good financial experience should be a personalised, intelligent service to help customers take advantage of the risks and opportunities in the market, not simply an interface to execute a trade.

I see 5 key ways to build next generation financial experiences for your customers.

1. Personalise Everything

Finance is a broadcast medium. Content and data wings it’s way from source to destination as it is created and is the same for everyone. Like radio before Spotify or TV before Netflix there is no concept of user preferences in how this information is created or delivered. This creates over 90% noise for the customer with information about instruments, events, topics, in the markets being unrelated to their interests.

Personalise as much of the information provided to your customers as possible - this will have a dramatic impact on how your customer engages with your platform and ensures risks and opportunities in the market are not lost in the noise of chronological broad spectrum market news.

I will talk in more detail about the types of personalisation that can be designed but initially information should be filtered for each user based on their instrument preferences such as an equity or asset class.

2. Curate and Generate

There is a real opportunity to sift through the information to find the story. Financial content is arguably the most dynamic, fast moving and noisy content ecosystems around, however new Natural Language Processing (NLP) and Large Language Models (LLM) make it possible to create a more effective information methodology by finding and summarising stories in the public, private and paid financial content and data sources.

By converting available data sources into far more consumable and enriched market stories you will have more success in engaging the customer in researching a trade. Personalising these more consumable stories will have a far greater ROI than endless chronological financial market commentary.

3. Behaviour Linked Models

Very effective personalisation and recommendation models are now available as open source or in platform on most Machine Learning Platforms such as Sagemaker, TensorFlow or Hugging Face. This will provide you with an excellent system for recommending an instrument to any given customer based on previous behaviour interactions such as a specific equity due to a Tweet interaction. However over time the list of recommended instruments will grow and the success of the model will begin to decline.

The key to improved recommender systems across all interfaces is to recognise that whats important to a customer at any given times depends of what market risks and opportunities are presenting.

I would recommend linking the weighting of an instrument to the following data inputs:

1. Market Behaviour - where price action such as change and volatility will change the relative ranking of an instrument for the customer.

2. Community Behaviour - where the interactions of other traders with instruments such as content or trades build momentum and trends which will change the relative ranking of an instrument.

3. Commentary Behaviour - where the mentions and stories coming from the financial markets commentary such as news and tweets impact the relative ranking of an instrument.

4. Entity Management

When you are opting to build a a true real time personalised Financial Experience from the content and data you have ingested, it is critical to build relationships between entities - this will have a significant impact on the depth and breadth of the experience you offer to your customers.

The more instruments, topics, events, people and products that you can build relationships between the more likely you will be able to curate and generate market risk and opportunity information for the customer, and therefore provide a more engaging and profitable Financial Experience.

5. Financial Markets AI Chat

When you bring the above strategies together into a comprehensive Financial Experience Platform, you will now have the capabilities within your stack to create a truly functional and feature rich financial markets AI Chat assistant. Currently most examples of AI Chat in finance are simply a more advanced search than true personal brokers. With some key upgrades Financial Chat will be game changing for retail customers.

The key capabilities for ground breaking financial chatbot are:

1. Summaries of all available market commentary & data so they are more consumable and suitable for message format.

2. Personalised for each customer so risks and opportunities are provided to the user rather than offered as advanced search - for example Generative AI morning and afternoon market roundups.

3. Entity Enrichment - where every chat has entities such as Instruments and Topics identified and functional for improved exploration or drill down.

4. Conversational Trading - where the engagement and informing from the chatbot can easily move into execution and monitoring - such as ‘if $NVDA moves above $400 buy 100 at £10 a point and close at $450.’

5. Education Assistant - conversational education provision and assessment is perfect for chat with support on how to place and manage specific instruments and markets helping the retail investor take advantage of risks and opportunities

There are huge opportunities for innovative brokers, banks and app first retail platforms to gain market share by working on the above strategies.

I can help you transform your product strategy and develop the technology capabilities needed for the next generation of financial experiences. Please get in touch if you want to discuss this and much more in detail at adam@personalisation.ai.

Cheers

Adam