As we roll out new functionality we are now documenting all use cases. Here is the first, importing and engineering data sets in the application.
Click here to read more.
It doesn't take long to realize that data is agnostic to source. Web logs look similar to retail transaction data. Survey data and customer profile data can be handled in the same way.
When building a data platform, finding a niche is hard because if data is source-agnostic, so is the platform. The platform we have built is a generalist product; point it at any data source or stream and it will be useful. What links all the data sources together is another generic concept - customers.
Generalist products are hard to sell, primarily because it is hard to find audience insights that help contextualize the problems the product can solve.
For the past five years of building the product we have found it is easier to write code than create a succinct product proposition. In recent weeks, some ideas have been crystalizing and we landed on this:
Knowledge Leaps is a customer data platform for storage, engineering, and analytics of all types of customer data.
Knowledge Leaps is a cloud-based data management platform that allows for collaborative analytics, data management and data-workflow management.
There will always be a plentiful supply of data scientists on-hand to perform hand-cut custom data science. For what most businesses requirements, the typical data scientist is over-skilled. Only other data scientists can understand their work and, importantly, only other data scientists can check their work.
What businesses require for most tasks are people with the data-engineering skills of data scientists and not necessarily their statistical skills or their understanding of a scientific-method of analysis.
Data engineering on a big scale is fraught with challenges. While Excel and Google Sheets can handle relatively large (~1mn row) datasets there is not really a similar software solution that allows easy visualization and manipulation of larger data sets. NoSQL / SQL-databases are required for super-scale data engineering, but this requires skills of the super-user. As 'data-is-the-new-oil' mantra makes its way into businesses, people will become exposed to a growing number datasets that are beyond the realm of the software available to them and, potentially, their skill sets.
At Knowledge Leaps we are building a platform solution for this future audience and these future use-cases.The core of the platform are two important features: Visual Data Engineering pipelines and Code-Free Data Science.
The applications of these features are endless; from building a customer data lake, or building a custom-data-pipeline for report generation or even creating simple-to-evaluate predictive models.
Benford's law is how the IRS/HMRC can tell if the information you submit on your tax filings is fraudulent. When people lie on their tax forms they tend to use random numbers, when really the number distribution should follow Benford's law. https://en.wikipedia.org/wiki/Benford%27s_law