Financial markets customer experiences suck

Most financial customer experiences are like listening to the radio when everyone has Spotify.

I have spent the last few years designing and building personalisation strategy and platforms for the financial markets and created my first financial website over 20 years ago. I am passionate about improving the customer experience of users in the financial markets, I specialise in real time personalised financial experiences.

There are limited examples of financial customer experiences that meet the expectations of customers. Despite a glut of new apps, platforms and products i believe there is much room for improvement in how we deliver financial markets customer experiences. This industry wide malaise could be as result of a range of strategic factors, ranging from the prevailing churn and burn business model, apathy, poor innovation and technology capability, regulation fears or possibly a lack of imagination. In any event financial experiences are stuck in a broadcast world when everyone is streaming.

Its not just the execution platforms falling short, I recently analysed a prominent retail trading content house in the US and analysis of user behaviour and content strategy revealed that more than 90% of the content surfaced on the website did not contain information that was related to what the visitors were interested in.

Financial experiences can and should be improved if they want to survive what is certain to be a period of extreme disruption.

Top 5 areas for improvement

Let’s have a look at the top 5 reasons financial customer experiences aren’t matching the needs of their existing or prospective customers.

1. Content & Data isn't relevant to user interests

2. Very little social content

3. Traditional analysis isn’t ‘smart’ enough

4. Recommendations needed (within regulation)

5. Market connections must improve

Cut the Noise!

I had a difficult conversation a couple of months ago regarding a personalisation model I had developed with the data science team, several of the executives had trouble explaining the model, what the features were and why it was important. The model be discussed dramatically improved the relevance of the financial markets for each user, no mean feat and dramatically underplayed when explained to the client.

Let's take our earlier example of 10% of all content being relevant to a visitor versus 100% of all content, data, recommendations, search, emails, trends, screeners, notifications being highly related to the demonstrated behaviours and interests of each user - the experience is chalk and cheese.

Can you imagine browsing the internet or social and seeing the same ads as everyone else or everyone having the same facebook and Twitter feeds. Personalisation makes a huge difference to how we consume the markets, if we reduce market noise we can better identify risks and opportunities and ultimately help our customers enter and exit trades.

Social Content is more engaging

A lot of people spend half their day on Twitter and Reddit and youtube reviewing the markets. Gen Z is an extreme example of this and it does’t find trading ideas from the traditional analyst content, rather Gen Z almost five times more likely to get financial advice from social media than people aged 41 or over.

There are some providers of social content to financial audiences and there are some examples of Twitter feeds in major brokerage platforms but by and large those social feeds are pretty limited to select accounts and not linked to preferences. User generated content (UGC) tends to get a pretty rough deal from compliance departments but the risks can be controlled through adequate offensive word filtering and only using selected accounts, the policy framework for this also needs careful consideration, particularly if provided by a third party provider. Brokerages who do use Twitter content will be a little apprehensive

Other social platforms such as Youtube, Reddit, Tik Tok, Discord and Twitch are the new frontiers of finfluence and trading communities where people share risks and opportunities as professionals and amateurs alike. Depending on the terms and conditions of the business api’s there is a lot of opportunity to surface valuable financial content to customers. Meme stocks haven’t died, they will continue to captivate a younger audience that seeks the thrill as much as the returns, there is also the us against them mentality which is a strong pull.

Make market analysis and commentary smarter

Traditional analyst content doesn’t get the same traction it used to. To some extent this is because customers don’t trust that content anymore and broker customer research does indicate this is a factor in why many brokers and reevaluating their analyst teams, i think its part of the us against them mentality the pervades the financial markets, especially the meme stock craze. However I think the content is valuable but it’s not ‘smart’ enough, what I mean by that it's flat, it doesn't help the user execute a trade, it simply informs of a technical level or fundamental change to be aware of. It gets written and is quickly forgotten.

Brokers should be looking to build smarter tools to automate the entry and exits for instruments where a good technical setup exists, such as a technical level or a classical chart formation being triggered. These tools would notify traders when the right conditions have been met or perhaps even propose a trade that can be setup directly from the content - this would dramatically improve the usefulness and engagement of a expensive form of content and link the analysis to the execution. Sure compliance will need some reassurance but I would use this tool to at least let me know when to review and entry or an exit.

Recommendations (lets call them suggestions;)

The amount of linguistic gymnastics i’ve had to perform when talking recommendations to brokerage desks and compliance teams - ‘we cant say recommendations’. yes I get it - I would never get technology to recommend a trade (regulatory nightmare, however for sure you should be recommending instruments to research for a possible trade to dramatically improve engagement.

If you are developing a recommender system for your brokerage you wont regret it as it will improve your understanding of your customers, engagement will improve, and trading volumes will increase depending on the quality of your system. I will cover personalisation and recommender systems in more detail in another article but please consider a couple of key elements:

• Build component specific recommenders as well as a universal model, so you have context for the user so that you can recommend instruments when the user inputs such as in Search or Watchlist

• Make sure your recommended instruments or content have market awareness. You might have hundreds of recommended entities (instruments, topics, content) for any give user - there importance to the user and therefore engagement potential will depend on real time market conditions such as price or volatility - market awareness or context here allows you to rank a recommendation at any give time. Super important point!

• Your recommendations should ideally be a blend of content and collaborative filtering, cohort behaviour might not be providing the breadth of financial instruments you need so rules based models such as more or less familiar instruments to a existing preference might be a better place to start - after all your goal is to judge a recommendation so reinforcing a echo chamber serves little purpose either for engagement or model optimisation\

Connect Instruments & Content

A key unlock for any financial experience is bringing to life the interconnected nature of the financial markets. This can be as simple as linking equities within a sub sector or peer or competitor group or as complex as advising on a hedging strategy for a particular trade. The benefits of better instrument relationships can be measured in increased engagement - users will follow from a piece of content talking about Tesla more easily to an article about battery production or lithium mining than unrelated content.

By building great relationships between instruments, people, places, products, economic events and topics you can add value to existing content and better leverage recommendations and personalisation putting more relevant, more related information in front of your customers. You could start with enriching your existing entity database - but the real benefits come from the rich relationships born from a financial knowledge graph - where connection between entities in content and data sources can develop organically. I will discuss knowledge graphs and market relationships in another article.

Personally I think there has been too much focus on the UX/UI elements and not enough focus on the overall strategy of providing a personalised financial experience - there isn't a need to gamify a trading platform when the markets are constantly providing engaging opportunities and risks from minute to minute. Take the worlds best content, curate it for each user based on what they like and what they should like, link instruments to all other elements of the markets and give everything context based on real time data such as user behaviour and price.

There is a lot to cover when building a modern financial experience. In the coming months I will attempt to enlighten the market on how you might go about building one and what the benefits to the customer and the financial institutions and brokerages are.

Cheers

Adam Howard

Founder