The actual keywords used to search technical resumes

How important are keywords on my technical resume? Which keywords should I use to stand out? Will my resume will show up at all? Will I be screened out automatically based on an absence of relevant keywords on my resume?

All of these are common questions and concerns for technology professionals writing their technical resume. Few are familiar with the actual behavior of either the humans who read resumes – engineering managers, HR professionals, or recruiters – or the computer systems that assist in the process – Applicant Tracking Systems (ATSs) and resume parsers. As a result, these worries, anxieties, and concerns cause a lot of stress in the resume writing process.

At Leet Resumes, where we write technical resumes for free for experienced engineers, it’s an extremely important topic for us to get right. Most information available on internet forums, Reddit, or Quora is disinformation – a mix of urban legends, wishful thinking, and personal guesswork. The anxieties fostered by this disinformation are unfortunate, but understandable – so much is at stake!

Because it’s important to get right, let’s review how keywords are actually used in searching and sorting resumes and applications. From these insights, we’ll conclude with recommendations for how to best utilize keywords on your own technical resume.

With regards to keywords, you should be most concerned with the humans involved, not the ATS, not the AI technology, not the automations. Because of the nature of how HR and recruiting decisions are made, and the unique preferences of each hiring manager, it’s best to start with the human’s desires and practices, and then follow up with how the technology supports or enables those preferences. 

The actual behavior patterns of the humans using resume search technology is pretty non-dramatic, and about as straightforward and sensible as you might guess if you’ve ever observed consumer interaction with high technology. It is a human professional, often HR or recruiting, sometimes the hiring manager herself, who conducts keywords searches and iteratively whittles down a result set to a size worth reviewing manually.

Despite expansive claims, AI and ML play a relatively small part in the review of technology resumes. While both are extremely powerful methodologies, they have become overused marketing buzzwords in certain sectors of the economy such as HR software. There is no obscure body of knowledge about optimizing your resume’s keywords, comparable to the intricacies of SEO best practices, for working with AI or ATSs. It just simply isn’t the case.

For example, one bit of bad advice occasionally found in internet forums is the importance of keyword stuffing – repeating high-value keywords on your resume in order to stand out, or to be listed top in the search results. Sometimes the advice extends to including these keywords in small white font on a white background, so the computer reads it but not the human.  Duplicitous deceptions such as these are not productive, helpful or successful because that’s not how humans search.  (Never mind that a properly-trained, properly-rewarded AI would learn to sniff out the stratagems used by poseurs attempting to fool the system.)

It’s also worth mentioning that automated screening should not be a concern for the experienced technology professional. While automated screening exists at the blue-collar level, and certain entry-level clerical jobs, for the most part screening is not automated in white-collar roles.  For technology professionals, in many ways at the apex of white-collar jobs in the 21st century, where the volume of applications rarely exceeds the ability of a human to screen them personally, automated screening is simply not a factor.

Instead, it is the humans involved, and the limitations of those humans, that should be most concerning to software engineers and technologists. The fact is, humans are really not very competent at sophisticated keyword screening. Especially for the HR professionals assisting an engineer manager, they lack information technology training and skills, and do not possess an extensive and sophisticated toolset for refining keyword searches. Typically, they’re most interested in getting to a result set that pleases their boss or client as quickly as possible, not performing an effective, deterministic, search process that avoids statistically significant false negatives.

Behaviorally, they tend to default to simple keyword searches. The keyword search paradigm is powerful, but lossy.  Your own natural experiments on Google, StackOverflow, or anywhere there is a search engine, teach you that search is a multi-step process of guessing and refining. With each iteration, you achieve a better understanding of the trade-offs between specificity and breadth. And with each iteration, you refine your assessment of whether the result set is sufficiently compact to merit your detailed review.

The same is true for resume search, and for resume searchers. The most common search terms from recruiters and hiring managers are simple, almost embarrassingly so: “engineer”, “director”, “devops”, “security”, “Java”, “Python”, “EC2”, “Apache”, etc. This might seem like a great disappointment relative to what you know technology is capable of doing. Having been in the industry twenty years, trust me, it’s frustrating for the people who build search software for recruiters, too!

The HR professional conducting a search may be relying on a job description for keyword suggestions, or restricting their search to a geographical area, or using past experiences that have turned out well for them. But in reviewing tens of thousands of searches by recruiters and hiring managers, simple searches are most common, and there is almost no prevalence of complex searches such as “used Python to build a scalable, databases-driven content management system for the aeronautics industry.” 

Complex keyword phrases and sentences such as these are simply not performed, and, when performed at all, are not productive. 

Instead, what we see is that when it comes to title or field, searchers tend to utilize common nouns. “Lead engineer” or “Vice President” tend to be more productive than more company-specific versions such as “Lead Engineer, Internal Database Tools” or “Vice President, Omaha Operations.” Similarly, “mobile” or “gaming” are more frequently searched for than terms that are brand-specific, such as “GTA 5” or “iOS Safari browser engineering.” Broader searches yield a larger result set that is also consistent with the intentions of the hiring or engineering manager.

Conversely, when it comes to skills and capabilities, it is the more specific proper nouns that are more productive. “Rust”, “MongoDB”, and “Ruby” are much more commonly searched than generic or common nouns such as “database management”, “programming language” or “web server.”  For skills and capabilities, the generic terms are more likely to return false positives from a resume database. And the more specific proper nouns carry more signal about the specific skills of that professional.  

This pattern - common nouns for titles and field, proper nouns for skills, capabilities, and competencies - yields important insights for writing for your own resume. At Leet Resumes, our advice for technology professionals’ resumes follows from these observations:

  • Get your past titles correct. Technologists will sometimes skip over past promotions, for example, only listing the final title in an eight-year job instead of demonstrating each promotion along the way. While not directly related to keyword search, getting this right supports the next point…

  • Include future titles. Your resume is an advertisement for a future position. After all, if you wanted to keep the title you currently have, you wouldn’t be updating your resume. As such, it’s important to indicate on your resume the three or four job titles you’d actually accept for your next role. That way, keyword searches for the job you want – “Software Engineer II” or “Lead Engineer” – will return your resume in the results.

  • All the things. Please include all of the technologies in which you are proficient on your resume. Too often, it will seem obvious to a technology professional that “if I know Django, I obviously know Python,” leading to the omission of an important technology proficiency. But to the HR professional assisting a hiring manager, it is not obvious at all, and they may miss you as a result. Every technology in which you have demonstrated competency should be on your resume.

  • Don’t forget databases! At Leet we’ve observed that, for whatever reason, technology professionals too often leave databases and data stores off their resumes. Please be explicit and include MariaDB, Redis, Neo4j and every other database and data store you are familiar with in your technologies list. Do not expect that readers will know that LAMP stack or MEAN stack imply a knowledge of MySQL or MongoDb. It’s best to be explicit.

  • Be exhaustive. There’s no need to resort to foolish tricks or silly ruses to get your resume noticed. But what is required is that you mine your past experience, in detail, for accurate and detailed descriptions of the accomplishments you’ve had, the work you’ve done, and the technologies, fields, activities, and capabilities you’ve developed an expertise in. It’s hard work to get all of this information into your bullet points, but it pays off.

In summary, creating your resume does not need to be a stressful, anxiety-producing effort. Worrying about keyword strategies will do you little good in the real world. There is no keyword strategy that is more effective than accurately and exhaustively describing the facts of your actual work experience, talents and skills.

With these tips, you’re on the path to creating a more effective resume for yourself, or having Leet Resumes write a technical resume for you for free.