Situation: Hiring the best possible people is very important to the success of any company. Incoming hires, not only affect performance but also can have an impact on culture, work ethic, internal development and company standards. To get the strongest and most effective new hires for a company, recruiters look for the most important qualities in candidates including leadership skills, problem-solving approach, motivational attitude, goal setting and tracking, and communication.
Challenge: Human qualities are often difficult to quantify and interviewers are left to their own opinions and “gut-feel” to make a hiring decision. For virtually every position that a company hires for there are rounds and rounds of interviews and multiple candidates to meet with. How can we make the hiring decision more objective? How can we get a more consistent evaluation across interviewers? How can we reduce the amount of time and speed up the data collection and evaluation process?
Traditional Solution: When companies evaluate a group of candidates, they typically schedule the candidates to meet with several interviewers one-on-one. Then, they collect input on each candidate by giving interviewers an evaluation form which contains the qualities (mentioned above) in a rubric, typically ranging from 1 (least desirable) to 5 (most desirable). The evaluation forms are picked up after all the candidates have met with all the interviewers and the lead recruiter complies the results and forms a consensus point of view of the candidate(s). There are several issues with this approach. First, when people rate using a rubric it’s very easy to lose consistency if there are a large number of items (ie. qualities and candidates) to rate, or if the rubric is complex or hard to follow. Second, in these types of environments participants can come to the process with very biased points of view. In many instances, we’ve seen at least one person that is either a “hard-grader” or “easy-grader.” A hard-grader is someone that rates everything (i.e person) at the bottom of the scale. An easy-grader is just the reverse. The hard and soft-graders scores can skew the results for the group. Finally, the traditional group interviewing process takes a lot of time. People must submit their input, then its usually entered into a spreadsheet and then the results are tabulated and shared. If there are multiple candidates, multiple interviewers, and multiple human qualities to evaluate, then all of this takes time to collect and compile.
Let us show you a better and faster way using GridRank!
Create An Interview Game
We created a game for an upcoming interview day. Seven evaluation factors were entered into a GridRank “game” and are shown in the image below, they are as follows: analytical skills, management experience, interpersonal skills, cultural fit, certifications, out of the box thinking, and related work experience. We also entered in the five candidates that we interviewed, they are Robert, Henry, Linda, Maureen, and Pat. We had two interviewers that day, Scott and Kyle. Scott and Kyle each received email invites to enter their evaluations into GridRank independently. In all, there were 70 evaluations, which is the product of 7 factors x 5 candidates x 2 interviewers.
Once the game was created, the interviewers received an invite to play the game. During the interview or shortly after the interview, each interviewer entered his input into GridRank for each candidate and each evaluation factor. The interviewer needs\ed only to position and adjust the candidates relative to each other, not to a rubric, on the GridRank game board. So is was fast, simple, and consistent.
Play This Game Now
If you want to play the actual game yourself, click here to play. After you play, come back here and we’ll discuss the calculations and the benefits of using GridRank.
View the Results for the Interview Game
At the end of the day, the lead recruiter brought up the final results in GridRank (see image below). Any interviewer that had not submitted his inputs was clearly visible in the dashboard — see the interviewers names and the green dots next to them; if their input was missing, these dots would be red. Since everyone has entered their input, the lead recruiter is viewing the summary of the results for all five candidates (see below). Pat is visibly at the bottom, but its a little hard to see who is at the top.
Next, the lead recruiter clicks the Results button in the lower right. The main page of Results appears (see image below).
The pie chart shows that four of the five candidates were pretty close at the top. By focusing on the data in the table, we can see the exact number of each candidate. For example, Robert scored the highest with an overall %Weight of 22.4 The lead recruiter might want to drill-down and investigate particular evaluation factors, such as Analytical Skills. To do so, he needs only select Analytical Skills from the Criteria drop-down. GridRank produces a table (see below) that shows the top candidate, Robert, had the lowest %Weight of 13.6. Which, when compared to everyone else, is quite low. As part of the follow-up with hiring, the lead recruiter might want to raise this as an issue to the people that Robert could be working for. Otherwise, Robert scored very high for six of the seven criteria, and therefore, he’s an excellent candidate. Henry would also be an excellent candidate because he scored very high overall (21.9), and his Analytical Skills %Weight is 24.2, which is second best.
GridRank is better than the Traditional Approach
Using GridRank over the traditional method offers three improvements – consistency, integrity, and speed.
- Consistency — In Grid Rank, interviewers placed the candidates relative to each other, not to a rubric. Since there is no rubric, there’s nothing an interviewer needs to remember or put into his head — so it’s faster. Second, because interviewers rank all the candidates in one window, consistency is never an issue.
- Integrity — In GridRank, if a hard-grader or soft-grader positioned all of their interviewees at the top or bottom of the game board, it would have become readily apparent that the person is trying to skew the evaluation. Second, anyone that places all of their options at the top or bottom (in the traditional method, this would be equivalent to giving all 5s or 1s to a person), the results are mathematically negated by GridRank; so it’s very hard to skew the results when you use GridRank.
- Speed — In GridRank, setting up the game with multiple evaluation criteria and multiple people is fast. Once the game is created, and people are invited, their input can be entered in quickly and the results can be viewed immediately. You don’t need a room, whiteboard, or spreadsheet. Any change to the game, like adding a candidate or removing evaluation criteria can be done instantly and players can immediately play it. None of the previous work is lost or has to be redone.