Earn Better AI

Feature tested: Content Quality

Result: Passed (5.5/10)

Verdict: Medium- Produced a mixed output with some bullets retained

Expected behavior: For a Data Analyst profile that included SQL, Python, and Power BI experience across two roles, Earn Better AI produced a mixed output. Some bullets retained outcome framing from the input — a few even preserved approximate metrics where they were present. Others collapsed into task statements: "Used Python to support data analysis workflows." The full output is in the PDF below — read every bullet and mark which ones you would submit unchanged.

Why it matters / Conclusion: Earn Better AI performs noticeably better when given structured resume data with clearly separated experience, skills, and metrics. The Data Analyst profile retained some outcome-oriented phrasing and contextual relevance.

Feature tested: ATS Optimisation

Result: Failed (4/10)

Verdict: Low-poor ATS-safe formatting with mid-tier keyword coverage and limited contextual vocabulary depth.

Expected behavior: Earn Better AI performed in the mid-range on Jobscan scoring — the lowest result recorded this cycle, below the top performers. The exact score and keyword gap breakdown are in the Jobscan screenshot below. Check it against the role description we used to run the scan before drawing conclusions about which gaps matter most for the roles you're targeting.

Why it matters / Conclusion: The resumes exported cleanly and passed ATS parsing without formatting corruption, but the keyword strategy lacked depth for competitive technical roles. Core technologies like SQL, Python, React, and Node.js appeared consistently, yet surrounding contextual terminology and domain-specific phrasing were thinner than stronger-performing tools. The Jobscan results reflect this directly: acceptable baseline optimisation, but insufficient semantic coverage for high-competition data and engineering applications.

Feature tested: Input Handling

Result: Partial (6.5/10)

Verdict: Medium-Flexible free-form input support

Expected behavior: Earn Better AI accepts both structured form input and free-form text paste, which separates it from tools that lock users into sequential field-by-field entry. The screenshot below shows what the unstructured input interface actually presented — the tool made an attempt to parse and redistribute paragraph content into the correct sections before generation began.

Why it matters / Conclusion: Earn Better AI handles unstructured input better than locked form-only resume builders by allowing direct paragraph pasting and automatic section redistribution. However, the parsing layer is not reliable enough to eliminate manual cleanup. Mis-assigned sections, incomplete redistribution, and formatting inconsistencies still required intervention before generation. The workflow reduces pre-processing time compared to rigid builders, but does not fully automate raw-input conversion.

Feature tested: Editing Burden

Result: Partial (5/10)

Verdict: Medium-High rewrite requirement with significant post-generation cleanup for competitive applications

Expected behavior: We estimated 40–55% total editing effort across both test profiles — higher than Wonsulting AI on the same inputs. The annotated output below marks every bullet we would rewrite before submitting — task statements flagged in one colour, missing quantification in another, keyword gaps in a third. The Data Analyst profile required less intervention than the Full Stack Developer output; that difference is visible in the markup.

Why it matters / Conclusion: The generated resumes were not submission-ready without meaningful editing. Across both profiles, a large percentage of bullets required rewriting for stronger achievement framing, quantification, ATS targeting, or clearer technical positioning. The Full Stack Developer output carried the heaviest editing load due to weaker contextual understanding from unstructured input. The tool provides a usable draft foundation, but users should expect substantial refinement before applying to serious roles.

Feature tested: Formatting

Result: Passed (6/10)

Verdict: Medium-Clean ATS-safe exports with solid visual structure

Expected behavior: Clean layout with readable font hierarchy and adequate white space. The exported PDF holds up at standard page margins and imports without visible corruption. One issue noted: section spacing behaved inconsistently when content volume was uneven — visible in the Full Stack Developer output where sparser sections created an unbalanced bottom half on the first page. The Data Analyst output, with more evenly distributed content across sections, exported cleaner

Why it matters / Conclusion: Formatting is one of Earn Better AI’s stronger areas. The PDFs exported cleanly, maintained readable hierarchy, and preserved ATS compatibility without corruption. Layout quality remained stable for balanced resumes like the Data Analyst profile, though uneven content distribution exposed spacing weaknesses in the Full Stack Developer version. Overall presentation is professional enough for real-world use, provided the underlying content is manually improved.