The following are tips and best practices for calibrating positions & how to leverage filters afterwards. Not every practice may be pertinent to every role that is calibrated, but these principles are meant to be holistic.
Guiding Principles
❏ It is important to remember that there is no 100% “correct” calibration. Calibrations may differ from even position to position based on what a hiring manager needs.
❏ Calibrations may not always result in match scores of 5. Some roles may only have match scores of 3 as the highest score based on an employer’s defined criteria and the availability of the skills in the existing talent pool.
❏ A candidate’s match score for a position can change as candidate profiles, including Ideal Candidate profiles, are refreshed. This typically does not result in a significant change in the match score, such as 2 to 4 or 4 to 2.
❏ It is not necessary nor recommended to use all of the available Calibration filter criteria. Using a lot of filters, you will find leads that match one criterion but not all of the criteria which will negatively impact the match score.
❏ After defining the calibration criteria, it is best practice to review some leads that are the highest and the lowest ranked to validate the overall direction of the calibration. You can review the first few using the calibration assistant. The calibration should be adjusted/edited only when you strongly disagree with the results.
❏ Filters on the position should be used and saved, to drill down into leads even more than calibration does. This helps to ensure that the widest net is being cast in calibration to help identify unique candidates for the role.
❏ Eightfold’s Best Practice for Calibration
○ Use Calibration as your intake guide- reviewing requirements with the hiring manager and capturing your notes in the calibration notes section
○ Location(s) of Job
○ At least 3 ideal candidates
○ At least 3 preferred skills & no more than 1 required skill
○ At least 3 alternate job titles
○ Add years of experience range
○ Avoid too many Must Haves or Must Not Haves
○ After completing the calibration components, go back to the Ideal Candidates and confirm your entries for Skills, Titles and Experience are still aligned.
○ Verify the calibration against some of Leads at both the highest and lowest matches to see if they align to the requirements. If not, revisit your calibration in the specific areas of concern and make adjustments.
Breaking Down the Components of Calibration
1. Ideal Candidates
Ideal candidates have the most impact on the match score and their skills and experience should set the framework for the rest of the calibration. Being a strong performer and having a diverse, yet complimentary background in skills and experience across your ideal candidates is a bonus.
When selecting Ideal Candidates
❏ Select at least 3, but no more than 5. Having more than 5 will make it more difficult to zero in on specific preferences and requirements for the position.
❏ Job titles for Ideal Candidates should be as close as possible to the name of the position you are calibrating. You can always modify a profile’s title if they have a non-traditional title.
❏ Consider current employees at your company.
❏ Ensure they have a rich profile that includes the skills and experience that meet the needs of the position. Usually these individuals are working in the role currently.
❏ If there are not enough internal ideal candidates, pull in from applicants, outside sources (like LinkedIn), and/or leads for the role. You can always add ideal candidates to the position later as you work through your various candidate sources.
❏ Select individuals who are:
❏ currently in the same or similar position
❏ have the skills and experience required for the position
❏ have a rich profile
❏ have been extended an offer for the position but were unable to join the organization for other reasons
❏ For roles that span multiple business sectors, choose an ideal candidate from each area. For example, Academy Consultants may have degrees and/or experience in Business, Healthcare or IT. All candidates will be matched against the same combination of all Ideal Candidates.
❏ If this is a new role in the organization that did not previously exist, it is recommended to look outside of the organization through sources like LinkedIn.
❏ Select individuals who are currently in the same or similar role who have the skills and experience required for the position and ensure they have a rich profile.
❏ If there are specific skills or experiences that you are looking for in a role, it is best to include an ideal candidate(s) that has this skill present in their profile.
❏ Do not select individuals who are no longer performing in the role/similar role. Doing so may introduce unwanted or inflated skills and experiences beyond the position's requirements and will have an impact on the match scores.
When to Change Ideal Candidates
❏ If the position remains open for a long period of time, periodically validate the ideal candidate group, to ensure each is still relevant given the ongoing profile refreshes.
❏ When copying a calibration from another position, validate the copied in ideal candidate selections to ensure they are relevant to the new position.
2. Refine requirements through Skills and Experience
The selection of skills and experience criteria should enhance or compliment the Ideal Candidate group. Adding in skills and experiences that are not related or that contradict those of the ideal candidate group may negatively impact the Leads pipeline and match scores. We recommend that you stay as broad as possible for calibration and then leverage filters afterwards.
In setting the Skills and Experience sections
❏ Focus on Skills, Titles and Years of Experience
❏ Use the ideal candidate group and job description to identify requirements as a starting point.
❏ Limit the number of Must Have and Must Not Have Skills and Titles. These are rarely used outside of certain industries, sectors and roles. You will have another opportunity to add in various combinations of skills through filters.
Skills/Keywords
❏ Maximum of 5 skills with a focus on the hard skills
❏ Review the related skills for each skill selected to validate the relevancy and adjacency to those selected. This will help identify potential outliers that can impact the calibration and matching.
For example, a recruiter is hiring for a Technical Support Analyst and is looking for someone with a tech background and skills of SQL, Troubleshooting, Computer Science and Customer Service. In reviewing the related skills of each, the first three skills share several related or adjacent tech-type skills. Troubleshooting also includes a related skill of Customer Service. However, when reviewing the Customer Service related skills, these are mostly retail and non-technical type skills. Including Customer Service as a skill will impact the match scores and potentially pull in Leads that do not align well overall with the position requirements. This is not a wrong calibration, but the recruiter will need to decide if this is an appropriate calibration. If Customer Service is retained as a skill, the system will learn this is a valid relationship within your company and continue to strengthen and improve the matching over time based on company’s hiring trends.
❏ Limit Required skills to 1, if possible
❏ If there is a specific skill that is a must have or is more important than the others, mark it required.
❏ Required or excluded skills will knock out Leads that do not have this skill
❏ Required skills will impact the Match Score of applicants missing this skill
❏ Preferred skills are equally weighted within the skill category of calibration. Required skills are given more weight (e.g., there are 3 preferred skills, each is weighted 20%). The majority of skills added should be preferred.
❏ Excluding a skill will deprioritize applicants and impacts the Match Score for those with the particular skill.
❏ There are times that grouping skills is recommended as it allows multiple skills within a group to be evaluated as one. This reduces skill dilution in the calculation group for.
❏ The criteria within a group and between the group and
individual skills should be “OR.” Related skills will then be inferred against each individual skill within the group.
❏ Putting quotes around the skills such as “Computer Science” would execute exact word matches.
3. Job Titles
❏ For an optimum calibration, add at least 3 alternative Titles for the role, but avoid generic titles (ie., Technical Expert).
❏ Avoid acronyms to provide the platform with more context
❏ Ideal candidates are good sources to suggest and validate related titles
❏ Choose titles commonly used, avoid titles exclusive to your organization
❏ Calibration does not look for specific title matches unless it is marked as required. Requiring specific titles is not recommended as organizations vary from company to company in the titling of positions and can be added as a filter.
4. Years of Experience
Relevant years of experience vs. total years of experience is often based on the role and level. In more technical or specialized and management roles that involve specific training, skills or certifications, relevant years of experience is more commonly used. Within non-specialized/non-technical and individual contributor roles total years of experience is more common.
❏ Years of experience will not exclude any candidates, but it does impact the match score. Having more years of experience than preferred has less of an impact on the match score than having not enough experience. You can always adjust this as a filter.
❏ If a job description specifies requirements to both total and relevant years, give preference to relevant years of experience.
5. Education
Degree indicates the highest level of attainment and will help to add or remove employees from the leads pipeline. Graduation year helps to further refine the results for interns and early career/entry level positions.
❏ For the Intern positions, select Bachelors and set the graduation years for the anticipated graduation date
❏ For Early Career positions that require a Bachelor's degree, select Bachelors and set the graduation dates accordingly
❏ Graduation year is in full year increments
❏ If Graduation year is set for the current year, leads have not yet graduated, but are anticipating graduating this year will be included
❏ If Graduation year is set for the end of the prior year, the leads anticipating graduating during the current year will be removed from the leas pipeline
❏ The leads pipeline will be filtered to include candidates with a Bachelor's degree awarded during the specified graduation period
❏ There is no impact on the match score for higher level achievement (e.g., Master’s)
❏ Validate graduation year from the profile/resume
6. Seniority
The seniority filter is mainly used with top level positions of CXO, VP and Director in specific industries and sectors. It can also be used for interns across industry and sector. For all other levels, it is recommended to not use this for initial calibrations.
❏ If after reviewing the matched leads, candidates are found to be too high or too low level, then the seniority filter can be applied and leads rechecked to see if that better aligns the matched candidates to the employer’s requirements.
❏ As a general guide, seniority levels are considered:
- Intern - An active student looking for opportunities
- Entry - Recent graduate or early in their career (1-2 years of experience)
- Mid-Level - Individual Contributor with some work experience but not as much as a senior member
- Senior - Individual Contributor with more experience than mid-level. This includes leads as well
- Manager - Professionals with the lower level as direct reports
- Director - Professionals with Managers as direct reports
- VP - Have high level management roles
- CXO - C Suite professional
Copy Calibration
After a position has been calibrated and the calibration verified and confirmed, it can be used as a starting point for the next similar position.
❏ Using the ‘Copy Calibration’ functionality a recruiter can quickly copy over choosing to replace or add-to the calibration of the new position
❏ All components of the calibration copy over, with the exception of the title, location, location preferences and years of experience. These will need to be set within the new position.
❏ All other calibration criteria should also be reviewed and validated at this time to ensure the framework is aligned with the new position’s requirements.
❏ Ideal Candidates
❏ Skills
❏ Titles
Filters
Once a calibration has been completed, the filters on the position will default to show those selected in calibration. Recruiters can always add more filters to further consolidate the list of leads presented. If calibration casts a wide net, then filters can help narrow down the list of leads to create a more ideal sourcing experience.
Available Standard Filters
Note that these are standard filters, but your company may configured additional filters which would result in additional filters visible in the pipeline.
❏ Highlights
❏ Strong Career Growth
❏ Top School
❏ Similar Industry
❏ Likely to Respond
❏ Location
❏ Job Code: add in previous positions that you want to search within for leads
❏ Skills/Keywords
❏ Able to add in additional skills/keywords and indicate if it is required to be have been recent experience
❏ Title
❏ Seniority Level
❏ Companies
❏ Able to select current companies only
❏ Experience
❏ Education
❏ ATS Details
❏ Last Application Date
❏ Last Rejection Date
❏ Last Contacted Date
❏ Campaigns: filter on Eightfold Campaigns
❏ Other Options
❏ Tags
❏ Include/Hide Applicants for this Position
❏ Personal Info: Email, Phone, GitHub, Notes
❏ Languages
❏ Visa Status
Saved Filters
If you create a filtered view that you’d like to reuse again, you can always select “Save view” which will provide an option to save the view for the current position or all positions worked on by the user. Recruiter can name it and then it will be visible under Views.
General Tips
❏ Eightfold will rank applicants based on their Match Score against the completed calibration but will never remove an applicant from a position based on their Match Score. Calibration and filters will narrow down the Lead pool and remove Leads if they do not match either criteria.
❏ Eightfold does not generate error messages. This is most notable when entering skills. When entering a skill, especially a grouped skill and you have unexpected results with your candidates (e.g., Leads drop from 12k to 600). Go back and make sure the skills text is formatted correctly.
❏ If the Leads pipeline is large (1K+), the pages do not all load at once rather they are indexed. You may notice that as you page back and forth and see Leads change on the pages until the indexing has loaded the full pipeline.