AI in Recruitment : Is Mumbai closer to Delhi than Agra?

Jobseekers prefer to work closer home, their native town or their current location. They may also prefer specific locations because there are more job opportunities in that city. For example, Mumbai is a hub for financial services and Bangalore for IT jobs. That said, IT companies now have centers across all major metros and even in small cities like Indore, Jaipur, Trivandrum.

Jobseekers are willing to move from (say) Agra to Delhi, however, it is hard for an organization to convince anyone to move from Delhi to Agra. Charm of a large metropolis, with its educational, health, entertainment and modern lifestyle, is attracting talent towards larger cities. It has become a one way street.

As a recruiter (and hiring manager), when I look at a candidate, is he more like to move to Mumbai from Delhi? or will he prefer to move to a location near Delhi, say Agra? Often, geographical distance does not represent the user preferences. Unless there is some personal connect with a smaller town or incentives are offered with a promise for better location in the future, candidates are unwilling to move to smaller city or town. (Note - Agra is also developing very fast, preferences can change in the future).

Location is a simple "Yes" or "No", yet there are many variables which come into play in the Indian context. Some of the jobseekers want to live close to family and some away from it.  And preferences evolve as "the family" evolves and needs of the family change. A large number of jobseekers are willing to change location for the "better opportunity".

Location Preference Within a City
Yet, we see several employees depart because Gurgaon or Noida are too far from their current residence. Within a city, geographical distance or the daily commute is a major driver for employee satisfaction. An employee who was unhappy with his daily commute may eventually change the city itself (and not change his residence within the city).

AI Algorithm Must Understand the Preferences
The nuances of large and small city, distance within the city and also, personal preferences are all challenges for the AI algorithm to overcome.

- Vivek Jain

Please also see my blog post on (1) AI in Recruitment - Understanding Skills and Designations, (2) Story of Naukri Job Alerts, and (3) AI in Recruitment - Do Job Descriptions Represent the Intent of the Recruiter? 

Story of Naukri Job Alerts

Naukri.com is the market leader among with career sites in India, with market share currently at 63%. Naukri.com has 30 million+ registered profiles and a large part of these registered members receive a job alert every alternate day or on a weekly basis. Job alerts only contain freshly posted jobs on Naukri.com in last two/three days. It is probably the main reason why Naukri Job Alerts have one of the highest open and click through rates. Yet, jobseekers complained of relevance of jobs sent. That was identified as one of the important problems to address in early 2010.

The process of improving the job alerts was an incremental one. We built the logic step-by-step and with every incremental step, our understanding of the relevance problem improved.

1. Discovery of “Role” – I tend to believe one major variable than we discovered was “Role”. A deep dive in the behavioral data showed several interesting patterns. Jobseekers were not sticking to their Functional Areas (departments) and were applying across Functional Areas.

a. Pattern of apply clearly indicated that Role was more important than functional area.

b. We had roles which were very similar present in multiple Functional Areas, for example, sales role existed in Industry specific functional areas. GM Accounts existed in Accounts Functional Areas as well as the Top Management Functional Area. Also, several functional areas were close to each other, for example, Accounts and Banking.

2. Limitation of Keyword search – Key skills entered by jobseekers represents what they consider as important. Logically, a search on jobs should use the key skills entered by the jobseeker. However, some of the jobseekers had not entered their key skills. A large gap existed in the key skills entered and their skills as apparent from the CV. We needed a robust mechanism which did not fail because of the data inadequacy.

3. Handling of Categorical Variables – When we compare two jobs and their relevance to the jobseeker, attributes like “role” were important. The key challenge was to translate this into a distance function that can be used in predicting relevance for the jobseeker. Similarly, attributes like Industry, Location required identification of a good distance function.

4. Jobseeker Resume – A match between a jobseeker’s expertise and the requirements from the recruiter is essentially a match between the CV/resume and the Job Description. Of course, there are challenges – if a CV is old or a job description is incomplete, this may not work very well. Yet, we needed a mechanism for matching the candidate CV and the job description.

5. Apply Behavior - It is very much possible that apply behavior of a jobseeker will deviate from the CV/resume.  Apply behavior can provide insight into asiprations of the jobseekers as well as help identify classification errors. Incorporating apply behavior in identifying matching jobs for jobseekers is another significant challenge.

Naukri Analytics team identified the above challenges and incrementally solved them in association with the product team and the technology team. And of course, we noticed a major improvement in relevance feedback from jobseekers.

We are not done on solving this technical challenge. Analytics team is looking to hire smart Data Scientists to join its rank and work on solving these – if you are interested, please click to here to apply.

Email Marketing – Keeping Customer's Profile Current

Email marketing has multiple objectives like every other business initiative. For example, email campaign may focus on encouraging customers to make transactions that are profitable and also, ensuring that the same set of customers come back again for their next shopping trip. However, today’s profit cannot be the only objective, ensuring customers return in the future is probably equally important.


What is the trade-off between the two? In my last post, I emphasized on how customer segmentation improves RoI of an email marketing campaign or engagement strategy. If a particular email is only focused on current set of deals on offer without ensuring that customer keeps the profile information current (email address, mobile number, location, preferences, like or dislikes), targeting of offers will sooner or later, start missing the direction.

As a marketer, you will miss the current likes and the current interests that have changed since the customer first shared them with you. With limited and probably incorrect information, message will lose relevance. Customer may start marking your emails as spam or simply delete them or unsubscribe.

What percentage of your customers come back through email? Is it less than 2%, or as high as 50%? Do customers lose interest over time and if yes, what is the rate of decay in the level of interest? It depends entirely on your product category.  Travel gets maximum response in and around holiday season or long weekends. Ecommerce is driven by seasons and in category like jobs, interest level dips over time with annual revival during appraisal time.  If a significant percentage of customers come back, can you slowdown the process of customers losing interest? Or by making them update their profiles, bring them back to the active pool.

Should an email marketing campaign necessarily contain information that urges the customer to update the preferences and the interests that you have. Probably yes…

Or can you do one better? Include this only for customers who have not updated their profile information for some time. For customers who have recently updated the profile information, they continue to get mailers focused on maximizing transactions.