How to Find a Company's Real Problems So You Can Address Them in Your Application
Generic applications address the public pitch. Standout applications address the real problem. The four research layers — and the sources behind each.
By Michael Robert · Co-founder
Most job descriptions read like a checklist of skills. Behind that checklist sits a specific problem the team needs solved, and companies rarely spell it out in the posting itself. The people whose applications stand out are usually the ones who found that problem on their own and wrote straight to it. Four layers get you there: what the company says in public, what the role was opened to fix, where it's losing, and what it ships against what it has chosen not to build.
Why doesn't the public version of a company tell you what matters?
Companies publish what they want the world to see: mission statements, press releases, "About Us" pages. What actually matters, the real problems underneath all of it, rarely makes the cut.
The gap is not an accident. A posting has to attract a pool big enough to fill the role, so it describes the work in terms broad enough that a hundred people can see themselves in it. The specific thing that broke last quarter, or the bet that just got funded, stays inside the company. Most applicants write back to the broad version, which is the version every other applicant is writing to as well.
What are the four layers of company-problem research?
Each layer asks a different question and draws on different sources. Run all four and the overlap tells you which problem is real; a finding that shows up in only one layer usually isn't.
- Earnings calls or investor updates (for public companies)
- Founder interviews on podcasts (for private companies)
- Product changelogs and recent launches
- Job postings across the company (where they're investing)
- Which problem the founder keeps coming back to, across more than one interview
- Where the money is going: which functions are hiring fastest, and which product lines got a launch and a press release in the same quarter
- Job description, read twice: once for the skills, once for the verbs. Every place it says fix, rebuild, scale, unblock or migrate is naming work in progress rather than a qualification
- LinkedIn profiles of people in this role at the company (what did they do?)
- LinkedIn profiles of the people this role would report to and sit next to, since the team around a job describes the job
- Recent org announcements or promotions
- Why this role opened now rather than six months ago
- What success would look like by month 3, and again by month 12
- Whether this is part of something larger: check product launches and hiring sprees in adjacent functions
- Analyst or press coverage naming the competitive set, plus any head-to-head pages the company publishes itself
- Customer complaints on Twitter, Reddit, or ProductHunt: wherever people say the product falls short
- G2 or Capterra comparison pages, where the review text names the competitor the buyer was choosing between
- Founder or CEO interviews where they discuss competition
- Which competitor takes share from them?
- In what scenario do prospects choose the competitor instead?
- What's the one thing the competitor does better?
- Public roadmap or changelog, whichever one the company actually keeps current
- Recent product launches
- Feature requests or missing features mentioned in reviews
- The company's integrations or app directory, where the products it connects to reveal the ones it never built itself
- What have they actually shipped in the last six months, and does it match what customers keep asking for?
- What's the most obvious feature missing from the product, and what would it cost them to build it?
A general company research pass before applying covers the first three of these. The fourth, reading what a company ships against what it has chosen not to ship, is how a company-level problem narrows to the one this specific role was opened to solve.
What does four-layer research look like in practice?
Growing the consumer app while the money moves upmarket
Block's two most recent earnings calls point the same way. On the Q4 2025 call, CFO Amrita Ahuja named "the pace at which we move upmarket" as a variable in guidance and said the company stays focused on connecting its ecosystems. On the Q1 2026 call, Nick Molnar said the product "is continuing to resonate more and more with larger upmarket sellers," and Owen Jennings described connecting "both sides of the counter," Square on the seller side and Cash App on the consumer side. The positioning is broader than a simple payment processor.
There is no real posting to work from, so this is the layer you would build yourself. Say the listing is a growth PM role and the previous PM was promoted internally. LinkedIn is where that shows up: who held the title, when they started, and whether they moved up inside the company or left it. A backfill behind an internal promotion means the job already existed and someone outgrew it; a brand-new headcount means the company decided this work needed its own owner. The start dates usually tell you which one you're reading.
Cash App competes with Venmo (peer-to-peer), Apple Cash (integrated), and PayPal (the category leader), and peer-to-peer is the seam where it has been ceding ground. That CNBC report covers the first quarter of 2025, when Venmo's revenue rose 20% year over year while Cash App's gross profit grew 10%, and Jack Dorsey told analysts on Block's own call, "I just don't think we were focused enough and had enough attention on the network and the network density, and that is our foundation." Block's answer is a network play, described in one fintech write-up as leveraging its capabilities to build networks that only Block can create by combining the Cash App customer base with Square's seller network. Neighborhoods on Cash App, which Block demoed at that investor day, is one experience built on that combination, offering participating sellers lower transaction fees and Cash App customers rewards for shopping locally.
Cash App has launched stock trading, while Block's FY2025 10-K puts invoicing, payroll and the rest of the business-tools layer inside Square rather than Cash App, and a review of Cash App Business lists accounting software integrations among the features rival processors offer and it does not. The same narrowing shows up in geography. Cash App shut down its UK service in September 2024, and Block told shareholders it was "focused on growing within the U.S., not expanding into new markets." Those gaps look like deliberate product choices.
Consider a hypothetical B2B SaaS recruiting company with no public financials to triangulate from.
Betting the roadmap on assessment
CEO interviews mention "candidate assessment is the bottleneck" and "hiring pipeline velocity matters." The positioning leans on speed and quality, ahead of price.
Their LinkedIn shows a VP of Product for Assessment hired three months ago. That hire points to assessment becoming the company's main product focus.
Competitors include a job board that added assessments, an ATS with an assessment module, and a scheduling-first tool. Each treats assessment as a supporting feature, which leaves the assessment-first position open.
Recent launches: assessment templates, video interview features. Not yet shipped: competitor intelligence, enterprise integrations, analytics. Every launch deepens the assessment product itself; nothing yet connects it to the systems a buyer already runs, which is what an enterprise buyer asks about early.
What if I don't have time to run all four layers?
How much time should I spend on this research per application?
Four hours across all four layers, and closer to five if the company is hard to read. Layer 1 (what they publicly care about) takes roughly 45 minutes. Layer 2 (the role's real job) takes another 45. Layer 3 (where they're losing) is the deepest at about 90 minutes, since comparison pages and review sites have to be read rather than skimmed. Layer 4 (what they're shipping versus not) takes an hour. That is half a day of work, so spend it selectively. Run all four layers on the roles you would actually take. On everything else run Layer 2 alone, which still gets you an opening line that names the specific work the team needs done.
What if the company is private and doesn't have earnings calls or analyst reports?
You triangulate the company's priorities from adjacent sources instead. Founder interviews on podcasts tend to be candid, because nobody runs a live answer past comms first. The case studies on their website will tell you which problems they actually solve for customers, and job postings across the company show you where the money and headcount are going. Even a CEO's LinkedIn posts are worth scanning, since people tend to write publicly about the thing they're most worried about.
What if I find conflicting signals in different sources?
Conflicting signals usually mean the company hasn't settled the argument internally yet. When what leadership talks about doesn't match what the product roadmap shows, that gap is real work you could help close, and it makes a strong cover-letter opening or a first question in an interview. Frame it as a question rather than a verdict: "Everything you've shipped since January is self-serve, but both leadership interviews this year are about enterprise. Which one is the roadmap actually funded against?"
How do I tell which of the listed responsibilities the team is actually failing at?
The four layers converge on it. A responsibility that also turns up in an earnings-call line, in who the team just hired, in a comparison page where a competitor keeps winning, or in a feature that has been missing from the changelog for a year is the one under real pressure. Write the application around that one as a single claim about the company's problem you can defend, and name the evidence that pointed you there.