INTRODUCTION

Many job seekers use AI resume development tools with the intention of providing shape and polish to their work history. For those with little experience in layout and design, AI resume development tools can help provide structure. However, many job seekers assume they are ready to launch a search with the output of these tools, giving them a false sense of security that the AI-generated resume will pass muster.

Our purpose was to assess how helpful AI resume development tools may be to job seekers. The resume is often a lightning rod topic and many job seekers will receive contradictory advice on this front. For those unsure which advice to follow, AI resume development tools are often presumed an authority on modern resumes.

For this report, we focused on the general job seeker using free AI resume development tools. This job seeker would not have in-depth understanding of modern resume best practices or various content strategies used to highlight candidacy and value offering. As such, the general job seeker would not be well equipped to discern which tools add value and which tools are obscuring a candidate’s value offering.

METHODOLOGY

The same resume was submitted to each AI resume development tool. The resume was submitted to 13 AI resume development tools: CareerFlow, Enhancv, Hired AI, Jobright AI, Kickresume, LockedIn AI, Managejobapplications.com, Promap, Rezi, Sheets, Swooped, Teal, and Zety. We tested free AI resume development tools to evaluate tools readily available to job seekers.

As a control variable, we also tested Jobscan’s non-AI resume development tool. This relies on data mining, pulling phrasing from forms and placing it into pre-assigned, formatted sections. Data-mining does not create fresh content for you. It instead populates fields based on your submissions.

Some AI resume development tools failed to generate a downloadable resume. Either the website crashed, it kept buffering and never created a document, the user could not download the file, or the site requested payment to access the file. For FinalroundAI.com and resume.io, the download button did not work. The person testing the tool would click the button and nothing would happen (repeatedly). Wonsulting kept crashing. At the time we tested, the link for myytrudy.com would not open. Career.io and t1resume.com required payment to access a download, so we did not include those in our study.

Responses were evaluated based on:

• Content structure: Clearly positioning the candidate for a target role, visual variety to make information easier to read and retain, and intuitive placement of sections.

• Language: Succinct, accurate, and easily-skimmed phrasing. Recognizing hiring authorities prioritize recent experience, resumes were also assessed for showcasing career momentum.

• Formatting: Points of emphasis (bold, color), font size, and style choices like infographics, icons, company branded elements, and more.

• Accuracy of the resume product: Deletion of valuable data and addition of erroneous or inaccurate information.

This research focused on free AI Resume Development Tools as opposed to Large Language Models, which were evaluated in a prior report.

ASSUMPTIONS

While there are many preferences in resume writing, some standards exist. For this assessment, the following are considered basic characteristics of a resume:

• Position the candidate for a specific function or field with a target title and summary.

• List work experience, including company names, titles, and dates (location optional). List experience in reverse chronological order.

• Reinforce qualifications and transferrable skills by listing keywords or skills in the summary and throughout the experience section.

• Metrics add weight and credibility to scope and accomplishments.

• Present information in a professional tone with strong verbs.

• Organize content in clearly labeled sections.

• Incorporate additional topics such as technical skills, languages, and credentials, if relevant to the role.

More nuances of resume writing can hone quality. With this in mind, we evaluated AI resume development tools based on the following:

• Visual variety makes information easier for humans to read and retain. (For example, paragraphs longer than six lines are less likely to be read, an all-bullets format creates a blurring uniformity, overuse of bold obscures rather than highlights, etc.).

• For the general job seeker, the education section should be after experience. An exception would be recent graduates with no experience.

• Single column resumes perform better when subjected to ATS and human readers, as information can be more easily accessed if it follows predictable sequence and placement.

• Resumes should be crafted to showcase career momentum, meaning more information in the most recent role and subsequently less information in each role to follow.

• Phrasing should reflect the medium. The use of “I, me, my, we, our” sounds like social media or LinkedIn, making the resume sound less professional.

• Additional features of resume language include concision, strong verbs, only a select few modifiers, and clear sentence structures catering to skimmability.

RESULTS 

Content Structure

Most AI resume development tools seem to agree we need a target title and summary, with the exception of Hired AI and Sheets. Teal hammered its resume intro down to one branding statement, a huge mistake for higher-level or complex searches. CareerFlow’s summary was pulled from LinkedIn, most likely due to it being primarily used for LinkedIn feedback as opposed to resume development. Unfortunately, CareerFlow does not seem to recognize that the resume and LinkedIn profile are different beasts.

Resume with prominent skills in the summary provide a clear outline for decision makers. Readers spend the most time and attention on the top third to top half of the first page on the resume, so decision makers want to know your skillset quickly. However, only JobrightAI and Managejobapplications.com included ample keywords in this sweet spot.

Another style element that improves reader experience is a mix of paragraphs and bullets, especially in the Experience section. This visual variety keeps the eye moving through the page and prevents overwhelm from a document comprised solely of bullets. Only CareerFlow, JobrightAI, and the control resume from JobScan (non-AI) offer this feature.

Almost all of the tools place Education after Experience, appropriate for everyone except recent college grads with no professional experience. This was the top-scoring category in the entire study.

Most AI resume development tools recognized that single column resumes perform better when subjected to sourcing and screening tools, as well as human readers. Readers need to be able to find what they’re looking for quickly and easily, yet double column resumes don’t offer predictability in content placement. Some ATS also don’t parse double-column submissions effectively.

Language

CareerFlow, Hired AI, LockedIn AI, Promap, Rezi, Swooped, Teal, and our control non-AI tool Jobscan flattened career progression with sections coming in around the same length. Kickresume, managejobapplications.com, and Zety churned out submitted content. Enhancv, Jobright AI, and Sheets seemed intentional about crafting work history with more information in the most recent role and subsequently less to indicate career growth visually.

Most of the tools recognize that the resume’s voice is different from social media. CareerFlow, Kickresume, Promap, and the control Jobscan seem to think the resume should sound like social media, including “I, me, my, we, our.” It has long been standard for resumes to remove personal pronouns (or related forms) for concision. Since primarily young, very green talent is using these personal pronouns in resumes, most job seekers should avoid them.

Several tools tried to fluff up the resume with modifiers: engaging, consistent, detailed, effective, results-oriented, effectively, seasoned, remarkable… While sometimes a strong adjective can help offset screen out factors or convey exceptional results, most often they just add noise. Modifiers in combination with AI’s tendency to drag out phrasing mean the output from most AI resume development tools will fail to pack a punch. Only the control non-AI resume development tool, CareerFlow, and to some extent Teal manage to avoid loosening up the language.

Sentence structures reflected submitted language to remained fairly simple. CareerFlow and Kickresume suffered from unnecessarily complex sentence structures, stacking clauses and conceptual interjections.

Formatting

Presumably, formatting would be among the top benefits of using AI resume development tools, yet they fall short here. A well-constructed resume uses color, boldface text, and bullets to draw attention to elements that establish candidacy (titles, credentials) and make the job seeker stand out from the competition (accomplishments, often with metrics). None of the AI resume development tools delivered content with strategic formatting.

Formatting — Color

Most resumes were black and white, which is fine. Black and white resumes can still use formatting effectively to draw the eye to key elements. When color was used, it was not used strategically.

Rezi and Promap AI developed the entire resume in a dark blue. Enhancv uses a teal column on the right side of the page, with stylized initials in a circle. Kickresume, LockedIn AI, Teal, and Zety used color on candidate name and section headers. Kickresume also accented dates. LockedIn AI also added color to the company name. Oddly, CareerFlow only added color to contact information.

In terms of look and feel of the document, Jobscan’s non-AI resume offered a nice mix of paragraphs and bullets, as did Jobright AI. CareerFlow also generated a mix of paragraphs and bullets; however, since it looks like it was pulled directly from LinkedIn, credit may go to the profile, not the tool. Teal and Swooped incorporated a visually pleasing balance of color and white space. Not strategic, but they look nice.

Jobscan, CareerFlow, Hired AI, managejobapplications.com, and Promap AI boldfaced the title over other details like company name, city, state, and dates. While other strategic highlights fell short on these tools, boldfacing job titles does help establish qualifications.

Formatting — Font Size

Enhancv generated teeny tiny 8pt Calibri font, which will send hiring managers searching for their readers and zooming for information. Kickresume and Promap come in a hair larger. Small, but more legible, font sizes were used in CareerFlow and JobrightAI. CareerFlow and Kickresume make matters worse due to tighter line spacing.

On the other end of the spectrum, we see large fonts and excessive white space from Hired AI, LockedIn AI, Sheets, and Zety. These just look sloppy. Rezi and Teal also have large fonts that teeter on the edge of appearing childish.

Formatting — Images, Icons, Infographics, and more

Speaking of childish, Zety and Enhancv add a Languages infographic. This takes up needless space and does not add value. Enhancv inserts cheesy diamonds prior to key accomplishments pulled into the teal column. Similarly, Kickresume, LockedIn AI, and Rezi use small icons before contact information. These style elements detract from more meaningful content.

Enhancv and Promap use the resume to brand their companies. Enhancv’s download has “Powered by Enhancv” and an Enhancv web address. Promap includes a link to “promap profile.” Not great for the job seeker.

Formatting — Selection and Placement of Client Data

CareerFlow uses Skills as an incredibly vague section header, with Relevant Skills and Top Skills as subheadings. These simply add clutter and confusion. Stronger subsections might be Technical Skills or Interpersonal Skills.

Jobscan lists the candidates street address, an old-school resume feature that can expose candidates to privacy concerns and age discrimination, since this is no longer common practice. The value of listing a GPA is also questionable, as it can subject candidates to prejudice.

Formatting — Page Dimensions

Most resume development tools we tested generated a downloadable resume with standard page size. However, Sheets does not convert or print smoothly. It also generates a file outside standard page sizes. It claims to be designed for U.S job seekers, so this is an interesting issue.

Accuracy of the Resume Product

Jobright AI leans heavily on submitted language, looking more like data mining than unharnessed Generative AI. At this stage, data-mining delivers more accurate content more closely aligned with candidate voice. It takes more liberties in the summary, where the AI phrasing is looser, adding empty lead-ins like known for, skilled in, and eager to.

CareerFlow, Hired AI, LockedIn AI, Sheets, and Teal substantively edit down submitted content, eliminating core qualifications, content, and wins. Most likely attempting concision, this often robs work history of distinguishing characteristics. LockedIn AI spit out the most egregious hack job, severely truncating the entire document. CareerFlow and Hired AI also throw in some hallucinations. HiredAI had creative mode ramped up way too high, generating 13 metrics that don’t exist.

Enhancv, Kickresume, and Managejobapplications.com rely heavily on data mining, yet still manage to weaken the content by removing metrics, turning verbs into nouns, weakening language (efforts, helping), and introducing wordiness (strong focus on, offers passion in).

Promap, Rezi, and Zety suffer from wild hallucinations that create much more work for job seekers to locate discrepancies, reinsert valuable wins, and correct inaccuracies. Most resume builders that unleash more creative AI also elongate phrasing, add fluffy modifiers, and introduce boilerplate like Promap’s “managing multitasking demands in fast-paced settings.”

Swooped combines the faults of overediting and hallucinations, cutting 11 metrics and delivering a whopping 23 false metrics and off-base keywords.

As the control, Jobscan’s resume development tool that does not employ AI relies heavily on data mining. It still falls into the trap of overediting, making significant cuts to the experience that eliminate metrics. JobScan also took all of the personality and skills out of the summary, reducing it to a single branding sentence.

The chat below breaks down the number of deleted metrics and sections in combination with instances of hallucinating metrics, keywords, or adding a strange section / line such as Zety’s addition of three nonsensical accomplishments under the education section.

BROAD FINDINGS

None of the AI resume development tools strike an effective balance between data-mining client submitted materials and elevating content through Generative AI. The weak spot here is Generative AI. When tools are calibrated for more creativity, we see output going off the rails.

A serious concern with AI resume development tools is that many job seekers will imagine AI is elevating the resume when it is eroding the quality and accuracy of the resume. Since the general job seeker is unaware of resume best practices, users of AI resume development tools often hit the market with a subpar resume.

Even when weighing document organization and formatting, many of these tools do not add value and fail to use style elements strategically. Formatting, bold, and color are used for visual appeal in some tools, but they do not draw the reader’s eye to elements that establish candidacy and highlight points of distinction.

Another risk is that with a dozen or so AI resume development tools on the market and several users, submissions will look the same as opposed to customized and distinctive.

At this stage, a resume development tool that relies on data mining or client submitted information will deliver a more accurate resume requiring fewer major revisions. The custom-created resume that was submitted to all tools outperforms both the data-mined and AI generated versions of the same content.

The strongest resumes will remain custom created, even when plain black and white, without slick formatting. Second best, data-mined content based on keyword-rich, metrics-infused, and client-approved submissions. Our study recommends job seekers avoid using resumes generated from AI resume development tools.

For the full report with sample resumes and test notes, please download the pdf: