Referrals 2.0: when job posts spread like Memes
A deep dive into the IHC model based on MIT-backed research - not about “a new algorithm,” but about how people behave when a job is not just a chance… but a gamified asset.
We usually think of recruitment as a process: it starts with a vacancy and ends with a hire.
But in the digital world, a job post is also an informational object that spreads through networks.
And every candidate? Not just a respondent - but an agent in a social game.
In 2025, a group of researchers introduced a new model that combines economic incentives with network dynamics in recruitment.
Not another hiring tool - but a new way to think about how people act when a vacancy enters their feed as a signal, not just an opportunity.
📖 What was studied?
Title: Incentivized Network Dynamics in Digital Job Recruitment
Authors: Blas Kolic, Iñaki Ucar, Manuel Cebrian, Rosa E. Lillo
Goal: To model how people decide whether to apply for a job or pass it on — when incentives are in place.
Key questions:
What kind of networks spread job info most effectively?
What types of incentives (money, status, points) work best?
How does behavior shift when the rules of the game change?
🔧 The IHC Model: How Does It Work?
Imagine LinkedIn, or your social platform of choice.
A job post reaches someone in the network. Now, they have two options:
📤 Forward it to someone else
✅ Apply (and stop the spread)
The cascade continues until someone applies - then it halts.
So the job post either flows through the network or is absorbed by a successful hire.
💰 What drives that decision?
Incentives - and recursive ones.
50% of the reward goes to the hired candidate
25% to the person who recommended them
12.5% to the one who recommended that person
... and so on.
This creates a transparent, ethical “referral chain” - similar to recursive bounty systems (like the DARPA Red Balloon Challenge).
🧪 How was the model tested?
Researchers compared two approaches:
🔵 Oracle (Direct Recommendations) - like LinkedIn: a recruiter reaches out to candidates directly.
🔴 Social (IHC) - the job post flows through a social network, powered by incentives.
They ran simulations on three real social networks 👇
🔍 Simulation results from the IHC model vs Oracle recommendation system (Kolic et al., 2025)
🌐 The networks tested:
🌱 Copenhagen-sms = small team or school
Everyone knows everyone. Great for direct reach. Not ideal for long chains.
🏫 Uni_email = mid-size company or university
Structured into clusters (teams/departments). Good for multi-step spread.
🌍 Twitter_15m = large-scale social platform
Massive, with hubs and influencers. One share = 1000 eyes.
Best fit for IHC-style viral recruiting.
📈 Reading the charts:
Success rate - how often a hire was made
Chain length - how many people the job passed through
Number of applicants - how many people applied
Connectivity - how well-connected people in the network are
👉 Bottom line: IHC wins in networks that support organic spread.
💼 How can you use this?
You’re a recruiter in a small team?
✅ Direct outreach works.
❌ Don’t rely on viral spread unless your network is well-connected.
You’re building a platform or HR tool?
✅ Introduce referral gamification:
Bonuses for shares
Visibility boosts for active referrers
Badges or roles like “Hiring Ally”
▶️ You’re designing a referral system?
✅ Go beyond "bonus for hired".
🎯 Reward the chain, not just the end.
(Example: 50% to the hire, 25% to the referrer, 12.5% to their referrer, etc.)
📌 What does this all mean?
Recruiting isn't just about job listings - it’s a network game.
People aren't just "candidates" - they’re agents who make choices.
And the best systems? The ones that reflect human behavior and social motivation.
💬 Final thought:
If you want your job post to work - stop hoping a recruiter will “find the one”.
Let the job flow through your network.
Because sometimes, the best candidate won’t apply -
…but their friend might share it with someone who does.
📎 Research paper (Feb 2025):
Incentivized Network Dynamics in Digital Job Recruitment
https://arxiv.org/abs/2410.09698