PeopleTicker Proprietary Technology

PeopleTicker's proprietary technology uses a combination of advanced conceptual search techniques and probabilistic latent semantic indexing (PLSI). As a result PeopleTicker is the only job search and reporting tool available today that provides accurate, dynamic data based on your unique job description, regardless of skill set or industry.

Conceptual search: Two people use the same term to describe a concept less than 20% of the time -- a problem known as "term mismatch." (Furnas, 1987). Concept searches solve this mismatch problem by returning results that relate to the same concept as the query word, rather than the specific word used.

Probabilistic latent semantic indexing (PLSI): PLSI is a statistical technique used for the analysis of two key types of mode and occurrence data. It has applications in information retrieval and filtering, natural language processing, machine learning from text, and related areas. It is generally regarded as a more principled statistical approach than standard latent semantic analysis.

PeopleTicker Algorithm at Work

Job Description vs. Title PeopleTicker technology is based on granular job descriptions and Pay rates, not just job titles -- so fine points are taken into consideration. Unlike current technologies, we don't group together similar titles with less than 10% pay differences, or simply divide annual salaries by 40 hours to get an accurate hourly rate for contractors. PeopleTicker reports back real job description and employment data, gathered from actual employment engagements.

Why the "one size fits all" approach doesn't work Survey-based compensation tools attempt to create universal or "one size fits all" models for job titles, job descriptions, and experience levels. This allows them to maximize the number of data points for each job title.

Unfortunately, results are:

  • Very subjective
  • Too general
  • Only apply to existing employee base

PeopleTicker recognizes that:

  • Effective rates and salary data cannot be based on standard job titles
  • Rates and salary data vary based on specific skill sets
  • Indexing live job market rather than historical data is infinitely more accurate
  • The career market is relatively unstructured due to varying jobs and applicant skill sets being recruited

Problems with salary survey approach: The following table summarizes some of the difficulties associated with the gathering and comparing of salary and rate information for full time and contingent workers using old salary survey approaches.

Reality Problem
Companies have their own unique job titles Survey based tools ask users to make a selection from a predefined list of job titles in order to return salary data. Companies must match their current employee population to the predefined list of job titles. This requires a great deal of subjectivity.
Companies have unique skill set definitions for titles In order to obtain the highest percentage of matches, survey based tools create very general job descriptions that map to a predefined list of job titles. Companies must decide where employees fit in 'best'. This 'best fit' approach is riddled with inaccuracies that drive cost up and quality down.
Every company has some level of internal banding skill sets There are several methods companies can use to define internal banding. Some companies have 3 experience levels (Junior, Mid, Senior), while others have 4 or 5. The problem is that one company's mid-level may be another company's senior level. Companies that employ salary survey methods mix and match to force adherence that does not reflect reality.