Recruitment Transformation
Requisition Fulfilment Probability
Business Issue
A lot of work is wasted on sourcing profiles for Requirements (Reqs) where
there is low probability of submission and submissions of profiles which have
low probability of hires
Phase 1
Model
Predictive model to provide the probability that a
submission will be made against a Req successfully
Outcome Expected
This actionable intelligence will help the recruiters focus
on the Reqs where the probability of successful submission is high reducing the
time and resource wasted on working on the Reqs with lower probability of
success
Features, which could be used
Skills (Required), Skills (Candidates), Job Title, Positions, Req Location, Req Type, Req Category, Company rate for the skill, Education Level, Years of Experience, Years in Consulting, Availability of Consultants in a particular area for a skill and average bill rate of the consultants in a particular area for a skill
Phase 2
Model
Probability that the submitted profile will go to the
interview stage
Outcome Expected
This actionable intelligence will help to skip the time or
efforts that is getting wasted by recruiters in trying to submit resume where
the probability of a resume going for interview is very low
Features, which could be used
Division, Skills (Required), Skills (Candidates), Job Title, Positions, Req Location, Req Type, Req Category, Company rate for the skill, Education Level, Years of Experience, Years in Consulting, Availability of Consultants in a particular area for a skill
Phase 3
Model
Probability of interviewed candidates getting selected
Outcome Expected
This actionable intelligence will help in planning for
coaching, onboarding, forecasting and account management
Features, which could be used
Division, Skills (Required), Skills (Candidates), Job Title, Positions, Req Location, Req Type, Req Category, Company rate for the skill, Education Level, Years in total Experience, Years in Consulting, Availability of Consultants in an area for a skill
Probability of a Consultant Leaving Within 90 Days of Joining
Business Issue
There are quite a few consultants who resign within 90 days
of start (this is different from selected candidates who accepted the offer not joining) leading to revenue loss.
Model
We want to analyze the candidate profiles and interactions
with the company and come up with a predictability model that can identify the
candidates with higher probability of leaving within 90 days of joining before
the resume submission step.
Outcome Expected
Maximize the topline and bottom-line contribution from each
of the consultants placed with the client
Features, which could be used
Skills (Required), Skills (Candidates), Job Title, Req
Location, Req Location (Preferred), Salary (Existing), Company rate for the skill, Education Level, Years of Experience, Years in Consulting, Demand for a
particular skill in a specific area, Social Media Listening
Candidate Joining Likelihood
Business Issue
After going through the entire process some candidates fail
to turn-up for the project, leading to loss of revenue and reputation.
Model
Build a model which is trained through historical data to
recognize the differences between a candidate that joined and one that was
reported as a bad delivery.
Outcome Expected
Reduce bad delivery and improve topline and bottomline.
Features, which could be used
Not Joining Status, Salary, Company rate for the skill, Exp Bill Rate,
Candidate_Skills, Req Skills, Candidate City, Candidate State, Req City, Req
State, Willing To Travel, Willing to Relocate, Willing To Telecommute,
Availability Date, Interview On, Contacted Date, Source,
Division, Employee Status, Exp Start Date, Act End Date, Last Contacted Date, Submitted
On, Location Preference
Funnel Leakage Analysis
At each level of the hiring process, candidates drop off
leading to a small percentage of submissions joining the project
Model
Actionable insights for business / account teams in terms of what
is driving leakage and what they should be focusing on to reduce leakages
Outcome Expected
Improve the percentage of submitted candidates joining the
project
Features, which could be used
Req Skills, Req City, Req State, Job Title, Exp Bill Rate, Division, Req Type, Positions, status, Hits, Company rate for the skill, Hires, Company rate for the skill, Not Joining Status, Position Title, Education Level, Skills, Salary, Location Preference, Years Experience, Years Consulting, Employment Type, Submissions
Business Issue
There is no objective visibility into the time that it would
take to fill a particular requirement. There are times when the time given by
the client is less what it would take us to fill the requirement. It would be
imprudent to spend time on that Req. Also, it helps to set the expectations
with the client so that there are fewer missed deadlines.
Model
Based on various parameters predict the time to fill a Req once the entry is made in the system.
Outcome Expected
It will give us the time that a requirement in a skill type
will take to be filled
·
This will help us keep the candidates in the pipeline
for specific skills, which are more in demand to cut down on the time taken to
fill
·
Make process and resource changes to fill the
‘hot’ requirements to faster and respond to more requirements
Features, which could be used
submitted on, Interview on, Contacted Date, Expected Start
Date, Expected End Date, Duration, Days, Skills, Close
Fit Prediction, Business Issue, Model, Outcome Expected
Business Issue
The candidates drop out of the recruitment process at
various levels due to various issues.
Model
Likelihood of the candidate joining the project and
completing the project
Outcome Expected
Help reduce bad deliveries and voluntary attrition
Features, which could be used
Labor rate, Company rate for the skill, Skills, Req Skills, State, City, Req
Loc, Req City, Location Preference, Source, Req Type, Quality of Hire
Close Fit Analysis
Business Issue
Candidates are rejected by the client at various stages of
the recruitment process due to fitment issues. Also, the candidates leaves the
projects or are made to leave the project by the client due to fitment issues.
Model
Based on the requirements the recruiters should get a score
for a profile whether the candidate will be selected and recruited for a
project and perform well on the project
Outcome Expected
Help work on the profiles that have the higher probability
of success on the project. Will help in reducing the involuntary and voluntary
attrition
Features, which could be used
Req Skills, Req City, Req State, Exp Bill Rate, Division, Company rate for the skill, Position Title, Education Level, Skills, Salary, Location
Preference, Years Experience
Quality of Hire
Business Issue
The bad performance of consultants can result in potential revenue
and reputation loss.
Model
Analyze the performance of the consultants after they are
hired (evaluate with monthly and quarterly feedback)
Outcome Expected
This will help in predicting the performance of the
consultants in the medium to long-term, plan training or counselling
interventions to improve performance where needed and reduce attrition
Features, which could be used
Internal monthly and quarterly evaluation reports from (not in the system yet)
No comments:
Post a Comment