Delivering Behavioral Alpha

This case study illustrates how Essentia Insight’s behavioral reporting and intelligent nudges enabled a fund manager to tackle cognitive bias, improve fund performance, and generate a 20x return on his initial $50k software investment.


  • This case study features an established, successful European fund management institution (anonymized here for client confidentiality reasons) that runs over $300 billion in assets.
  • The portfolio manager, who we’ll call Karl, has over 20 years of investment experience and an active investment philosophy based on the premise that long-term outperformance can be achieved through a combination of fundamental research, active security selection and robust risk management.
  • Prior to working with Essentia, Karl’s $10bn Global Equity fund experienced two years of relative underperformance. Keen to address this and to build a more quantitative understanding of what was both helping and hurting his fund’s performance, Karl started using Essentia Insight in 2014.


Why outsource to Essentia?

  • To improve performance, Karl wanted to make greater use of the historical trade and position data he was already capturing. He also had an academic interest in behavioral finance and wanted to see how he could apply its principles to his everyday investment activity.
  • Existing technology within the organization was limited to traditional OMS, PMS, RMS and risk systems. An internal cost-benefit analysis carried out by the firm had already shown that it was better to outsource the development of any new technology, rather than try to build and maintain it in-house.
  • After a review of available third party solutions, Karl and his CIO selected Essentia. An important factor in their assessment was the strong decision-science pedigree of the Essentia team. Another positive was Essentia software’s ability to adapt flexibly to the firm’s investment style and process. But ultimately they chose Essentia because we understood their needs better than anyone else.


What did Essentia Insight reveal?

  • Essentia Insight is the analytics engine within Essentia’s data-driven feedback loop. It provides a powerful performance and behavioral overlay to the investment process.
  • Previous performance attribution analysis had shown Karl that his stock selection was not adding value, but it hadn’t provided any real insights into what he should do to improve.
  • After importing Karl’s trade, portfolio and benchmark data into Essentia Insight, it quickly became clear that fund performance was being damaged by the portfolio manager’s tendency to hold his losing positions for too long.
  • Holding loss-making investments for too long, whilst selling profitable positions too soon, is called the Disposition Effect. It’s a cognitive bias that is common to both professional and amateur investors and can have meaningful performance effects*.
  • Insight measured the frequency and impact of this bias on Karl’s performance through a comprehensive exit analysis. This included:
    • Running in excess of 1,000 scenarios or simulations, deriving not only the impact of each exit decision, but also an objective score of the quality of these decisions.
    • Measuring the average and distribution of the exit decisions across the portfolio for the sample time period and comparing this with the benchmark’s performance in the run-up to, and in the period just after, each exit.
    • Slicing the data by a long list of contextual factors and fundamental attributes, including sector, price momentum, day of the week, and holding period.
  • This analysis revealed that most unhelpful to Karl’s performance were stocks which had experienced drawdowns relative to the index and then did not recover significantly over consecutive weeks. In Karl’s case, these were typically small overweight positions which, whilst seemingly harmless in the context of the portfolio, were quite damaging when viewed in aggregate.
  • Insight went even further and showed that, by exiting these losing positions a month earlier, Karl would historically have saved an average of 3.1% ROI on the capital employed in each position.
Average Exit Pattern

ABOVE: Graph of Karl’s Average Exit Pattern. On average, he holds losing positions for too long (blue line), exiting at the near-term bottom.

From Insight to action

  • A popular and powerful feature of Insight’s feedback loop is the ability to set customized behavioral nudges. These can be used to alert investors to developments in their portfolios and/or any behavioral patterns that may affect their performance, before it’s too late.
  • In their regular review meeting, Karl and his Essentia Insight Partner (herself a former fund manager), considered the results of the analysis and decided that a behavioral nudge would be the best way to counter the impact of the cognitive bias identified by the software.
  • Using parameters defined by Karl, the Essentia Insight Partner configured a nudge which Karl called the Vulnerable Positions List (VPL). Generated on a weekly basis to coincide with Karl’s investment team meeting, this automated report would list those positions that were exhibiting the same price characteristics as those which had ended up hurting Karl in the past.
  • This VPL report quickly became valuable in helping to set the agenda of the weekly portfolio review. In cases where a drawdown alert had been triggered but the investment team was unanimous in its continuing support of the position, no action would be taken. Where there was dissent amongst the team, the holding was reduced.

‘Insight’s analytical depth and intelligent reporting means a fund manager can be more sure that he’s looking at the right things, at the right time. As a result, he can be more confident in his process, and more deliberate in his investment decisions.’

Greg Wallace, Essentia Head of Research

Process Takeaways

  • Karl’s use of Insight illustrates how Essentia can be used to reinforce an existing investment approach by bringing new, actionable data to the decision-making process. In this case, Insight data was used to nudge Karl so that he and his team were prompted to revisit their conviction levels and test their investment hypotheses in a deliberate and disciplined way.
  • Over the same period, data from Insight clearly showed that Karl and his team had demonstrated skill in deciding not to reduce certain positions that had appeared on the VPL. This data has allowed the investment team to evidence the success and rigour of their investment process when communicating with clients and investment consultants.
  • In our example of the VPL report, Karl set as 60 days the duration of any drawdown which would trigger a position alert. This value, as well as the other parameters behind the alert, is entirely flexible. It suited Karl because he runs a low turnover fund and he felt that a 60 day time horizon was sufficient for filtering out any short term volatility.  More frequent, shorter-term investors would, however, be more likely to define a shorter trigger period.
  • Other nudges and alerts can be created easily to meet the needs of the investment team using it. Insight supports watch lists for securities being monitored and not yet held. Insight also makes it possible to create what-if scenarios and run alerts retrospectively to see how performance would have been affected by different filters or decisions.

Investment Performance Outcomes

  • Over the last seven months of using the VPL nudge, Karl acted 15 times on the back of the alerts he received. Those actions directly led him to preserve $1m in alpha that would otherwise have been lost.
  • This $1m of alpha generated through a single nudge represents a 20x return on his $50k software investment.
  • Over the last 12 months of using Essentia Insight more broadly, Karl has made other performance gains: As well being braver about cutting losers, he’s also learned from Essentia that he can make more money by building his positions quicker, even within the confines of market liquidity.
  • A year after integrating Insight into his investment process, Karl had reversed his two year underperformance and was up 3.5% versus his benchmark.

UPDATE: How is Karl using Insight now?

  • Having started by using Insight’s VPL nudge on his large active positions only, Karl is now using it on all his active positions.
  • He’s also become more confident in exiting losing positions. We’ve been  able to show him that if he had completely cut the positions contained on the VPL report – rather than just reducing them – he would have added a total of 53 bps of alpha to his performance in the following six months. Moreover, if he had leveraged his team’s skill in deciding which positions not to touch, but cut the rest to zero, he would have made even more than 53 bps.
  • Karl has also set up a new nudge which alerts him to any stale positions in his portfolio. A position is flagged as stale whenever Karl has neither traded it, nor recorded any entries in Essentia’s investment journal (Essentia Note) for a predetermined period of time. To deliver the nudge, Insight serves up a selection of these flagged positions to him every week, to add to the agenda for his team meeting.
Trade Activity Visualization in Essentia Note

ABOVE: Trade Activity Visualization in Essentia Note. Karl uses Essentia’s online investment journal to record and update the rationale for his investment decisions. He has now configured Insight to alert him to positions that are categorized as stale. 

How does this case study relate to you?

  • Does your organization leverage investment big data to strengthen its investment process?
  • Do you know how much alpha you lose through the impact of behavioral factors and cognitive bias?
  • Are you still dependent on legacy technology, or is your team working with the next-generation technology now available to professional investors?

To explore these issues and see how Essentia software can help your investment team, contact us on or click here to arrange a quick demo.

* Research into the Disposition Effect found that winners sold outperformed losers that were retained by an average excess return of 3.4% per annum. (Source: ‘Are Investors Reluctant to Realise Their Losses?, Odean, T., The Journal of Finance, LIII(5), 1775-1798 – (1998)).