How can we help you?

Economists are notoriously ridiculed for their universal answer to almost any question - well, it depends.

When asked by a large customer service organization in a Fortune 250 company to provide them with predictive analytics to drive improvements to their operating efficiency and customer experience, the data team responded with a polite, yet sincere, “well, maybe”.

This is neither an atypical nor flippant response to a sincere request for help. It really does depend … Really.

• Firstly, there is the question of the data itself - what is there, where is it saved, how can it be accessed, what is the quality, is it reliable and consistent, etc.

• Secondly, is the specific data solution you are requesting really the one you need?

• Thirdly, would the data on hand be sufficient to support the outcomes you are hoping to achieve?

In this specific case, the answer to the second question turned out to be a resounding “no”. In fact, what this group needed more than anything at that moment in time was not advanced predictive modeling, but good, timely, easy to access, and fairly basic descriptive statistical insights. Counts and averages. Easy access to some basic statistics packaged into an interactive dashboard revolutionized their business within three months.

A start-up was investing heavily into developing its product - content for a digital health offering focusing on a niche within the mental health space. Despite their deep domaine expertise, they wanted to be sure they weren’t drinking too much of their own Kool-aid and that the topics they were addressing were, in fact really relevant for their target audience, so they turned to us for help.

Being a startup, resources had to be managed conservatively, so a large scale, survey-based research program was not the answer. Instead, we leveraged very scaleable web-scraping techniques to mine various public web forums where folks frequently go to seek advice from the wisdom of the crowd. In this way we were able to mine tens of thousands of real discussions, without violating any privacy considerations, and mined those conversations using natural language processing (NLP) techniques to model and quantify the range of topics actually being discussed by the target audience.

The resulting insights not only provided the statistically meaningful validation the client needed, but also revealed areas of opportunity for further product development, leading to an even more robust offering, more satisfied users, and more secure investors.

Skedastic has developed a number of fixed-price packages to aid with early-stage planning, fundraising, and launch aspects of getting a new project up and running. Please visit the Start-up page for more details, or reach out to us to discuss your specific needs, even if that just means a bit of brainstorming to help you figure out what those needs might be - we get it.