We are permanently monitoring technological breakthroughs, consumer behavior trends and regulatory changes in order to pick promising investment themes at an early stage.

The themes that are selected must be disruptive by nature in order to offer high growth rates over secular investment horizons.​

In collaboration with our network of contacts (private equity investors, start-ups, industry experts), we then perform a top-down fundamental analysis to determine the maturity level and hence the growth potential of the themes under consideration.

A segmentation into sub-themes is also realized, each sub-theme being assessed and then further segmented into very specific pockets in order to have a high allocation granularity.

This step is key in our initial top-down process because it will be used to monitor the themes’ evolutions in order to optimally position our portfolios over the themes’ different life-cycle phases.

The topological framework generated by the top-down process is then used to fill every themes’ buckets with publicly listed equities. At this stage, the only criteria that candidates must fulfill is thematic purity. This means that the selected companies must derive a major part of their revenues from services and/or products in direct link to the considered themes.

This large stock universe is quantitatively tracked and used as a thematic dashboard.

Our bottom-up process then continues by the reduction of the global data set into an investable universe of an average of 150 stocks per theme. Here, we skip the candidates that do not respect our risk management boundaries, mainly in terms of market capitalization size and trading liquidity volume.

Our proprietary multi-factor selection tool helps us to rank this investable universe by using the companies’ revenue growth rates, the trends in their operating cash flow margins and R&D budgets just to name a few factor examples.

Finally, the best candidates are going through a financial (DCF, balance sheet quality, sub-theme ratios…), competitive and technological analysis in order to build concentrated portfolios holding between 20 and 40 names.

The portfolios’ allocations are initially set based on our convictions. In addition, we also use the output of machine learning algorithms to guide us in the portfolios’ final risk optimization.

The resulting concentrated and conviction-driven investment vehicles are then rebalanced according to the themes’ macro and micro news flow as well as to respect our risk management guidelines (position size limits).