Updated: Apr 11, 2020
The hardest data challenge is to understand the investigative mind of an analyst or business on competitive landscape and make a predictive model. Building a competitive landscape predictive model without having human expert in the loop is almost impossible. Having human in the loop has speed and cost implication.
Why it is such a challenge to build a competitor relationship in algorithmic way?
The Transitivity Axiom
Transitivity of preferences is a fundamental principle shared by most major contemporary rational, prescriptive, and descriptive models of decision making. To have transitive preferences, a person, group, or society that prefers choice option x to y and y to z must prefer x to z.
Example: Deliveroo is a competitor of JustEat and Uber (UberEats). Uber is the competitor of Olo and Lyft. Deliveroo and Lyft are not competitors.
Asymmetric competitor relationships
What one company considers the other as competitor is not reciprocated equally. This asymmetry is driven by strategic direction, product/customer focus, geography, growth phase, resources and capabilities of individual company.
Employing domain experts to augment prediction will not solve the challenge in total.
Firstly it is cost prohibitive and difficult to scale when trying to do this for millions of companies.
Secondly it requires domain knowledge in individual sectors/industries.
Finally businesses are transient and time value of such information degrade rapidly. To act on these changes are impossible using human curators.
This challenge can be addressed in a constructive way that balances the expectation of the company without loosing the context.
CompX sources, validates, aggregates, productise and deliver Data that provides good, meaningful data that companies seek. This data is unique to them, tells a story, and is timely and specific.
CompX discovery infrastructure delivers us hundreds of data opportunities every month. Our platform has become a magnet for data assets that have never been exposed to private companies.
CompX data science team, empowered by proprietary technology and domain expertise, assesses every dataset we find for predictive power, reliability and compliance. The very best of these datasets are added to our platform.
Rich insights on curated data are derived by our expert team. Every nuggets of information are stored and analysed to expand the knowledge horizon.
Both the raw data and insights are structured and packaged for consumption by our customers.
Organization, documentation, expert support and quality control are paramount.
We offer institutional-grade reliability and deliver data in formats preferred by our customers.