Skip to main content

Discovery (Currently in BETA)


The Discovery feature will scan your candidate database and talent communities (if applicable) to surface and suggest strong candidates for your job opening. The list of recommended candidates can be reviewed using the dynamically created 'stacks' to further segregate the potential new candidates based on certain attributes and added to the job for consideration.

Where can I find this feature?

The SmartAssistant Discovery feature can be access in a number of ways. When you view a job listing you should see both a tab called "Discover" and there is also a button next to the Add Candidate button to the right of the candidate search box with the label "Discover Candidates". Both these links will take you to the Discover feature. If you cannot see either of these, please check with your administrator to make sure the feature is enabled and your account has been assigned access.


How does it work?

SmartAssistant will look into the company’s talent pools and past applicants alongside current applicants to identify who might be a good fit for this job using the MatchScore functionality. They will be automatically segmented into relevant groups based on job requirements and CV information. Recruiters will be able to see these groups and proactively filter this candidate pool by outstanding characteristics, such as suitable skill sets, nearby locations, relevant career experience, and more. These groups are called 'stacks' and will appear directly above the list of candidates. The idea is to provide an overview of the entire candidate pool, beyond just the applicants for this role, at a glance which makes the tasks of sourcing potential new candidates and shortlisting them faster and more insightful.


Recruiters can use the dynamically created 'Stacks' alongside conventional filters on the left to focus in on the listing of displayed candidates. Clicking on a candidate opens a sliding panel with their summary and access to their full profile pages so Recruiters can take further actions, such as moving them across application steps or inviting them to interview.


FAQ - Access and Processing

Q: Where do recommended candidates come from? 

A: We use the set of candidates who previously applied to jobs from your company as a source for recommendations. Among these, we only show candidates that are relevant enough for the job at hand.

Q: Will I be able to see all recommended candidates? 

A: No, SmartAssistant will respect the access groups the logged in user has and only display candidates which they have the right to see. This means that candidates visibility is granted by access groups, job shares or application shares. If I only had access to UK candidates then I wouldn’t see any recommended US candidates. SmartAssistant will still review the entire candidate population behind the scenes therefore a recruiter with less restricted access will most likely see more recommendations.

Q: Will I be able to see recommended candidates from Talent Communities/SmartCRM? 

A: Candidates from Talent Communities can be surfaced as recommended candidates but only if the logged in user already has access to see the Talent Communities that a candidate might be highlighted from.

Q: What triggers SmartAssistant to start looking for recommended candidates

A: For a job to be eligible for recommended candidates it must have a job posting but that posting doesn't have to be published status. The system will trigger the search for recommended candidates for newly created jobs once a day and the time it will take for them to appear will depend on how many new jobs there are in the system which has to be processed at the same time. 

Q: What triggers SmartAssistant to refresh the entire list of recommended candidates

A: As the Job Posting is a vital input into the calculation, any modfication to it requires a refresh to take the new/updated/removed posting information into account.

Q: Does SmartAssistant work with all types of job

A: Yes, recommendations will be provided regardless of whether the job is Public, Internal or Private

Q: Why is the Discovery feature still in BETA

A: Our Data Science team are constantly looking for way to improve both the quality of the candidate recommendations and the speed in which they are calculated. We want to make significant in both these areas before we would consider the feature no longer a BETA product.

FAQ - Stacks

Q: How are stacks created? 

A: Our Data Science team is able to extract a wide range of information related to employment, education, skills and others by applying Machine Learning and Natural Language Processing techniques on both job descriptions and candidate profiles. This information, along with historical data about interviewing and hiring decisions, is used to identify characteristics that are relevant for the job at hand. For instance, these can be “having a master degree in marketing”, “being located in the same city”, or “being skilled in a specific subject”.

Q: Can I select multiple stacks at the same time? 

A: Yes. Clicking on multiple stacks will provide an intersection, thus showing only candidates that are present in all selected stacks.

Q: Can I specify additional requirements regardless of the suggested stacks? 

A: Filters can be used for this purpose. For instance, if the candidate to be hired has to have a specific education degree, you can filter by such major and only candidates satisfying this constraint will be shown.