How the Programmatic Approach Is Revolutionizing Recruiting

About 150 years ago, the brilliant and successful merchant John Wanamaker said, “Half of my advertising budget is wasted; I just don’t know which half.” John was a visionary who invented things we take for granted today like the price tag and the “money-back guarantee.”

John noticed that at least half of the ads he bought appeared in front of the wrong audience. And for 140 years that was what marketers had to live with: Pet food ads were displayed to people with no pets and car insurance ads appeared in front of people who didn’t have a vehicle.

There are many similarities between marketing and recruiting, and if John were with us today as a recruiter or head of talent, he’d probably say, “Over 90% of the money I spend on recruiting is wasted; the trouble is I don’t know which.”

CPC is at the heart of the problem.

Our industry currently runs on the cost per click (CPC) business model. The problem is that these pricing models have no notion of quality or fit. When you pay for clicks, that’s what you get—tons and tons of clicks, with no accounting for their quality.

In a typical recruiting marketing campaign, if you spend $100 at $1 CPC, you get 100 clicks. From those 100 clicks, 20 people apply. You spend an hour reviewing those applicants just to find 4 or 5 who are qualified. In this scenario, employers receive hundreds of unqualified candidates, spend substantial time reviewing them, and pay for each one. As a result, employers are only willing to spend a small amount per click.

Introducing the programmatic approach.

The word “programmatic” gets thrown around a lot, but what does it really mean? In a programmatic system, a machine is programmed to make decisions that maximize some objective that was set by a human. And because machines can process more data—and make decisions more quickly—than people, they often can do a much better job, at a much lower cost.

In advertising, programmatic systems use huge quantities of data about individual users’ demographics and preferences to decide automatically which ads to show to which users in just a few milliseconds. As an advertiser, you can place your ad only in front of relevant people.

Because systems like these provide superior results to human-optimized campaigns, the programmatic approach took advertising by storm. It grew from zero to $30B in less than 10 years, transforming an industry that was based on relationships to one that is very efficient and data driven.

What programmatic means for recruiting.

In the next 10 years, programmatic is going to transform recruiting. It will make qualified talent available to you, anytime, on demand, while cutting weeks from the hiring process. And it will make the job search experience more human and rewarding, allowing candidates to quickly connect with interested potential employers.

Qualifications are the often ignored part of the job description, but they’re also objective and can be assessed accurately and automatically with artificial intelligence. Using qualifications as the language, I can set actions and prices, guaranteeing that for every position I get only qualified candidates at prices I set in advance. This might sound complicated and futuristic, but it’s already happening today. Let’s see it in action.

Meet Megan. Megan is looking for a new job. She’s visiting her favorite job site and uploading her résumé. The site, after a simple integration, is using a programmatic technology partner to extract the merits profile from Megan’s résumé. The profile includes detailed information about her work experience, industry, hard skills, soft skills, and education. The profile is then sent to employers with matching requirements. The employers, using a simple tool, set the bid for this type of candidate. Megan is then presented with open positions from the most interested companies. She selects the companies she’s interested in, and the employers receive an “Interested and Qualified” candidate.

It’s only the beginning.

We are still in the early days of the programmatic recruiting revolution. This is an exciting time to be a recruiter. We’re about to see a major shift. The next stage of programmatic recruiters will use smart tools to design requirements and find the perfect candidates. They will make a real human connection with high-caliber candidates rather than spending time on those who are irrelevant and unqualified—and this approach will have a tangible impact on their companies.

So what are you waiting for?

Leave the CPC inefficiencies behind and join the programmatic revolution. You won’t miss the piles of unqualified résumés—I promise.

The Secret to Identifying Skills That Aren’t Listed in a Resume

We’re all familiar with the job description—it’s one of the most essential tools in the recruiter’s toolkit. This versatile document helps you to come to an agreement with your hiring manager about what they’re looking for, publicize a role to the outside world, and get your hiring team aligned on a candidate’s suitability for a role.

A typical job description includes things like the required qualifications and experiences, as well as a list of desired skills the candidate will possess. In some cases, these skills are listed in the external job description, and in some cases, they may be listed internally in the company’s Applicant Tracking System (ATS). But one place that you will almost never see these skills listed out explicitly is on the candidate’s résumé.

There are a number of reasons for this: In the case where a company maintains the list of skills privately within their ATS, candidates won’t know which of their skills are most relevant. And even when the skills are listed in the job description, candidates may not always think to write their skills out on their résumé, either because they’re not sure if this information is relevant, they’re trying to save space with what they consider to be the most important information, or they may not have even thought of describing their abilities in this way.

But skills do matter to employers, and that’s why we’ve created a way for recruiters to easily uncover a candidate’s skills—even when they haven’t been explicitly spelled out in a résumé. Here’s how it works.

We used a large set of data, 6 million job descriptions and the skills that employers asked for when hiring for certain positions along with 50 million résumés, and built a broad coverage model that connects companies, positions, and skills. We trained the model on job descriptions, but it can now be applied to résumés to create an accurate portrait of a candidate’s hard and soft skills.

When a candidate uploads their résumé or LinkedIn profile to our system, it will automatically generate a list of skills. Some of these skills may be explicitly listed on the candidate’s résumé, but some of them are educated guesses based on our model.

We like to think of this model as a really smart career counselor who has vast knowledge about different roles and industries. As a job seeker, you might go to this career counselor and say, “I’m not really sure what my strengths are,” and she would tell you, “Well, you worked as a developer at Microsoft, so you’re adept at problem solving and collaborating towards a shared goal.” Then you would say, “Oh, that’s true! I do have those skills.”

For candidates, this is a huge win. They don’t need to make any changes to their existing résumé. Based on the information contained within their résumé, our system will generate a list of hard and soft skills. This includes things like technology or coding languages as well as soft skills like communication and attention to detail. Candidates have the ability to edit this list as well, to adjust their level of proficiency with skills or delete skills they don’t want to list.

By providing this list of suggested skills, we can help candidates better understand their strengths. It’s been shown that certain groups tend to over-report or under-report their abilities, so by proactively creating a list of suggested skills, we can help put candidates from different backgrounds on a more even playing field.

Recruiters can see a comparison between what they asked for and which skills the candidate has, which makes it easy for them to assess candidates on relevant criteria. It’s much more meaningful for a recruiter to know whether someone can handle a lot of tight deadlines than where they went to school, for example. This also gives recruiters a much more complete picture of the candidates they’re evaluating, and increases the likelihood of moving qualified candidates through the pipeline.

We’re excited about how this technology can better connect candidates and recruiters. Candidates will have a better idea of their strengths and find more relevant roles and recruiters will find candidates who are a better fit for the roles they’re looking to fill.
Want to see for yourself? If you’re a candidate, you can upload your résumé here. If you’re a recruiter, you can get in touch to request a demo here.

How programmatic is enabling recruiters to make visible impacts on their company

Recently, Uncommon got together with Greenhouse and Susanna Frazier, Head of Strategic Sourcing, Talent Acquisition at Universal Music Group, to discuss how programmatic is shaping the future of recruiting.

Listen for expert tips on how to:
  • Increase your pipeline with interested and qualified applicants
  • Decrease your spend on advertising open positions
  • Reduce time wasted on reviewing unqualified candidates

Uncommon Announces the ‘Talent Forecaster’, The Industry’s First Predictive Analytics for Talent

Anyone who has ever undertaken the challenging and tedious task of writing a job description knows that it’s tempting to imagine the perfect candidate while drafting your posting:  a Rhodes scholar, with a Ph.D. in Artificial Intelligence, and 10 years of experience with Tensorflow.

Unfortunately, this “purple squirrel” likely does not exist (for one, Tensorflow was only released 2 years ago!), and asking for those qualifications — or any qualifications far beyond what is realistic — is a recipe for failure. In most cases, however, hiring managers only figure this out after the job is listed, as hiring managers and recruiters have no idea how common different qualifications are in the population of real job seekers,  and thus how useful it is to list them in a job description.

Including too many qualifications, or qualifications that are too strict,  will result in nobody applying for the listing. With too few or too broad requirements — you’ll have an influx of low-quality applicants, which you’ll then have to weed through to find the few potential matches.

But how do you know when your qualifications are too high or too low?

That’s where the Uncommon Talent Forecaster comes in. The Uncommon Talent Forecaster is the recruiting industry’s first tool that helps you optimize your qualifications so that you can find the right talent in the shortest amount of time.

The tool is easy to use, allowing users to select qualifications and automatically determine the exact number of qualified applicants in their region, all in real-time. For instance, if someone asks for something rare – a PhD in Data Science for example – the forecast drops way down into the red zone, indicating there are too few people meeting your asks to obtain a solid talent pool for that role.   

Ask for something commonplace – knowledge of Microsoft Powerpoint for example, and the forecast shows you that there are thousands of matches, meaning you’ll be flooded with applicants unless you raise the bar. Ask for something somewhat selective, like Bachelor’s in Computer Science, and you see that you’ll get a reasonable number of qualified candidates — creating a way to ‘Goldilocks’ your hiring. And if you’re curious to see who those people are that meet your qualifications, you can just click a button to see a few sample resumes of some real people.

Check out the Talent Forecaster in action:

Uncommon GIF (1)

How did we do it? It all starts with data. Uncommon has a database of over 50 million resumes from across all industries, roles, seniority levels, as well as different parts of the U.S. Our proprietary models consumed each of those resumes to understand each person’s skill set, including years of job experience, industry experience, seniority, education background, and hard and soft skills.

Almost magically, as you’re entering your job requirements, we scan through all of those resumes, to figure out how many of the people near you would actually be qualified according to those requirements, and to tell you whether you need to make your qualifications more or less restrictive. And we do it all in less than a second.

Uncommon’s founder Amir Ashkenazi and I are no strangers to predictive analytics. At our last company,, we developed the advertising industry’s first supply forecasting tool to show ad buyers available inventory in a given marketplace. Today, that type of forecasting has become commonplace, with most major ad platforms, including Google AdWords and Facebook Ad Manager, offering such tools.

Now we’re bringing that same ease of use to the recruiting space with the Talent Supply Forecaster.

Have you ever had trouble creating the perfect job description? Tell us what happened in the comments below.

Uncommon Gets a “Rousing Applause” on The Chad & Cheese Podcast

This week, our CEO, Teg Grenager, was put through the gauntlet by two of the industry’s greatest tire kickers, Joel Cheesman and Chad Sowash on their show, “The Chad & Cheese Podcast”.

Memorable quotes by the hosts include:

Man, this product is so in depth…You’re pulling in programmatic components, you’re pulling in sourcing components and automation. Your pricing model is very interesting…And I’m really excited to see where this goes in the next 12 months.” – Joel Cheesman

It’s different, it’s something they [recruiters] can understand and it’s something that they can measure, because you are giving them exactly interested and qualified [candidates]. You tell me what qualified looks like, we’re going to give you exactly what that looks like…So from my standpoint, its definitely a big applause as well.” – Chad Sowash

Listen here:


Celebrating Our Launch in Vegas!

It’s hard to believe that we were in Las Vegas just a week ago, celebrating our launch at SourceCon 2018. It was instantly clear as to why this event has so many devotees that return year after year to soak up the innovations, wisdom and excitement from our industry.

Our team brought back countless amounts of new insights thanks to the sessions, keynotes and of course, the folks that stopped to chat at our booth about sourcing methods, pain points and what recruiters are wanting from their tech in 2018.

Thanks to ERE Media for putting on such a productive, valuable and lively three days! We welcomed a ton of new friends into the Uncommon community, and we’re excited to continue learning from all the thought leaders and sourcing wizards that we connected with.

Here are a few photo booth pics from the SourceCon kick-off party that we had the pleasure of hosting on the first night:

Follow us on Twitter to see more friendly faces throughout the next week!

How AI is Becoming a Recruiter’s Best Friend

Are you terrified you’re going to be replaced by a robot? This fear seems common these days, especially among recruiters. Many people working in talent acquisition are both curious about artificial intelligence (AI) – it ranked as one of the top global recruiting trends in a recent LinkedIn report – and worried about what it means for the future of their profession.

The concern is understandable. AI is an emerging field that is not well understood by the general population. However, I believe that AI is not here to replace recruiters. It’s here to make our lives easier, to automate tedious tasks, and to free up time so we can focus on things that really matter, like strategizing and building relationships with candidates.

In order to put some AI fears to rest, I’d like to look at three of the ways that recruiters can use AI to become more effective in their jobs:

1. Using AI to Write More Effective Job Descriptions

Writing a job post that attracts a diverse pool of qualified candidates requires more than a snappy company description and a rundown of job duties. In fact, your job ads could be unintentionally repelling the candidates you want, even if it looks like all the basics are there. Your descriptions could be too wordy, too jargon-y, too long, or even too short! Worse yet, you might be writing biased job posts without even realizing it.

How can you know if the language in your job posts is going to be effective and unbiased? Companies like Textio give you access to data crowdsourced from millions of job posts to predict the performance of your job post and help you improve it. Tools like these are nearly as easy to use as spell checkers, and the more posts they analyze, they smarter they get.

2. Using AI to Eliminate Scheduling Headaches

One of the most time-consuming and thankless tasks recruiters handle is scheduling interviews with candidates. The process generally takes multiple rounds of email – and hours of recruiters’ time – to finalize a date and time for an interview. All this back and forth introduces many opportunities for error. And don’t even get me started about what happens when someone has to cancel or reschedule!

This is why some of the latest scheduling AI tools like My Ally and GoodTime are so exciting. My Ally uses a virtual assistant, cc’d on your emails, to take care of scheduling, while GoodTime syncs information from calendars to find available times and make sure interviewers don’t get overbooked.

3. Using AI to Assess Candidates Efficiently and Objectively

In today’s market, recruiters are flooded with resumes, most of which come from unqualified applicants. There aren’t enough hours in the day to thoroughly assess each candidate, and unconscious bias can quietly guide even the most astute recruiters to choosing candidates for the wrong reasons.

Fortunately, resume screening is exactly the kind of tedious work with which AI can help. Moreover, since these AI recruiting solutions remove the element of subjectivity from the selection process, they can also reduce bias and help you build a diverse pool of talent. There are plenty of solutions available, like Pomato, which helps recruiters analyze candidates’ resumes, and HireVue, which integrates AI with video interviews.

While these point solutions can be quite helpful, other solutions fit AI into your current workflow. That’s the design philosophy behind Uncommon IQ, a qualification-based talent marketplace that sends streams of vetted candidates directly to recruiters’ applicant tracking systems or inboxes.

AI and the Future of Recruiting

AI doesn’t eliminate the need for recruiters. Rather, it allows recruiters to focus more on hiring great talent and less on tedious, low-value activities like digging through unqualified resumes. Hiring people is one of the most rewarding and beneficial parts of building a business, and I am genuinely excited about how AI can help us become more effective recruiters.