Reasonable notice periods may appear to be difficult to predict. Cases that may appear to be similar in terms of the Bardal factors can yield widely variable outcomes. Wouldn’t it be nice if there was a tool that could provide precise and accurate predictions based on relevant case law? Employment Foresight is the first software to utilize the power of machine learning to predict reasonable notice awards.
How does Employment Foresight differ from existing tools?
The predictive value of existing tools is limited. They do not take into account all relevant factors that courts consider or they rely on simple averages of court awards. Employment Foresight leverages the power of machine learning, using more than 20 relevant factors in a highly nuanced manner. Our research lawyers map the facts of existing cases, translating unstructured court rulings into structured data. Our program finds hidden patterns and the algorithm examines how judges have weighed the various factors in actual cases. Employment Foresight is highly accurate. On average, our predictions are within 8% of the actual notice period awarded in a case.
Consider our predictions of 10 very recent cases. To be clear, these are cases that Employment Foresight had never seen before. Our program had not been trained on these cases. For comparison, we also ran these cases through four other reasonable notice tools that are available online or for purchase. Note the differences in the predicted notice periods.
Employment Foresight correctly predicted 9 of these 10 recent cases, and got within half a month in the 10th case. For each prediction, we give a range of just 1 or 1.5 months. Other available tools provide ranges that are either excessively wide or incorrect.
Let’s look at two of these recent cases in a little more depth:
Case Study 1: Papp v. Stokes Economic Consulting Inc., 2017 ONSC 2357
Adam Papp, 32, worked for Stokes Economic Consulting for 1.5 years before being terminated without cause. He worked as a staff economist, a specialized position.
How much notice would a court award? If you take a rule of thumb approach, one would expect a notice period of 1 to 2 months. If you examined precedents from Ontario where the dismissed worker was 30 to 35 years old and had 1 to 2 years of service, the range would be from 3 to 9 months. Other online predictive tools give ranges that are excessively wide (ranges of up to 6 months) or provide answers that are inaccurate.
Employment Foresight predicted a notice period of 4 to 5 months. The court awarded 4 months.
Our prediction took into account the Bardal factors as well as the fact that the employer here made a number of negative statements in a reference letter. Neither the rule of thumb nor simple averages will uncover the patterns in the data that allowed Employment Foresight to arrive at an accurate prediction.
Case Study 2: Welch v. Ricoh Canada Inc., 2017 NSSC 174
Larry Welch, 46, and Kent Carroll, 58, were technicians with 26 and 25 years of service respectively. They were terminated without cause upon Ricoh’s loss of a service contract for 400 office machines. While Welch and Carroll did not have broad enough IT or computer skills to successfully pursue employment as service technicians in the IT industry, it was determined that their customer service skills and work ethic would give them options for subsequent employment.
What notice period would a court award to these dismissed employees? Other online tools give ranges that are excessively wide (ranges of up to 11 months) or provide narrower answers that are inaccurate. If you took a rule of thumb approach based on years of service, one might expect notice periods in the range of 14 to 26 months for both employees. Such a wide range is unhelpful.
For Welch: Employment Foresight predicted a notice period of 16.5 to 18 months. The court awarded 16 months.
For Carroll: Employment Foresight predicted a notice period of 18 to 20 months. The court awarded 18 months.
Other tools were largely unable to distinguish between Welch and Carroll and gave the same range for both employees. Our predictions took into account the Bardal factors as well as the fact that the employer here was in financial difficulty. Neither the rule of thumb nor using the average of awards in similar cases will uncover the patterns in the data that allowed Employment Foresight to arrive at these accurate predictions.
Employment Foresight leverages the power of machine learning to provide precise and accurate predictions driven by comprehensive data.