Aikido: Custom Strategy

For an ordinary person, investing is challenging and investing algorithmically is overly abstract.

Everyday people start investing to full-fill a financial goal. this could be generating additional income, preparing for retirement savings or overcoming financial challenges, including inflation, low interest rates, and wage stagnation. People care more about the outcome of investing rather than the process. However, it is not easy to achieve that outcome which as many users often stated "I know what I want but I don't know how to get there". They wanted a tool that would eliminate the effort and knowledge required to invest and allow them to spend time on other aspects of life.

Users wanted a personalised investing solution that would 'do the hard part for them'

While interviewing and testing other products related to prebuilt investment strategies that Aikido provided, users often struggled to in the decision of what strategy to choose, they would request the ability to 'blend' multiple strategies or a need to change a setting on a strategy. For more experienced investors, they were frustrated with the lack of freedom with current strategies Aikido provided. they wanted an environment where they could experiment with a strategy's settings and build their own strategy from scratch.To further define the needs we interviewed our users to better understand their investment goals:

1. What is the best risk vs return for me?
2. How can I invest to acheive my goals
3. I don't know where to start and don't know where to learn
4. I want something that will to most of the work for me with out the knowledge
5. I want to learn by doing

‍

The goal was to create a tool that would allow beginner and experienced investors to create test and deploy an algorithmic investment strategy tailored to their individual goals and needs.

I started design by ideating with developers and the investment team. This was done to maintain cross-collaboration and define technical feasibility and functionality from the get go.

The design emphasised optimising this process by minimising the number of clicks to complete input and providing smart defaults to reduce redundant actions.

For the product to perform, it had to provide many options to the user, with actions and steps repeated iteratively. The design emphasised optimising this process by minimizing the number of clicks to complete an input and providing smart defaults to reduce redundant actions.

The layout was designed to follow a similar structure of existing Aikido strategies for recognition and consistency. This information was contained on one page, as the user might modify multiple elements non-linearly. To filter for companies, users could select and define individual financial metrics and also select from a range of presets (sub-factors) designed to find stocks that met specific characteristics. This aimed to accommodate both experienced and beginner investors. It would be more accurate to filter stocks based on how a company performs relative to its industry competitors. This would allow searching for the best-performing companies relative to their competitors rather than the entire market stock market. The ability to add a metric and define its value based on a percentile relative to the industry average was added to achieve this.

‍

Prototypes were tested on 5 experienced and 5 inexperienced investors Excited about filtering based on industry percentiles, Liked the ability to sort the results Liked the concept of selecting sub-factors, as it ‘took out the difficult work' but struggled to understand which one to choose.

They were confused by the combination of sub-factors and individual metrics, and were unsure how the two were related. It was unclear to the user how factors worked the could not figure out which factors suited their goal. They found the number of options to be overwhelming. Users were confused about the difference between sorting and filters. They wanted more numerical descriptions for sub-factors and more detailed test data. Based on this feedback we improved the explanation of different steps, naming of steps was updated to better fit user mental models, strategy settings and results were separated into different tabs. This was well received by users they were quicker to fill knowledge gaps using tooltips and explainer features however, there still needed more clarity about the sub-factors primarily, how they worked, and what goal they would achieve. Beginner users struggled to relate a sub-factor with a given goal. In the end, was decided to remove sub-factors and in place, allow the user to choose a prebuilt strategy as a template. This approach afforded better results, as there was already data surrounding risk vs return and a given strategy's performance allowing a more informed decision to be made by the end user.

‍