In May 2025, I beat over 2,000 applicants to land a six-figure remote Data Scientist position. I had no referrals, no connections at the company, no Ivy League degree or FAANG on my resume. I cold-applied with three years of experience at a smaller private company you’ve never heard of — a role I also got by cold applying, without knowing anyone there either.

I’m not saying this to brag. I’m saying it because every piece of advice online about how to get a job without connections eventually tells you to go make connections, and if you’re in the middle of a job search right now, that advice is useless.

Why “just network” is bad advice when you need a job now

Open any article about getting hired without connections and you’ll get some version of the same thing: “Attend networking events.” “Build relationships on LinkedIn.” “Join professional organizations.” “Reach out to people in your target company for informational interviews.”

This is not wrong. Networking works, but it takes a long time. There’s nothing more awkward than going to a networking event or going to meet people with the sole purpose of trying to find a job — for all parties involved. Good networking is basically making friends, and that requires time and low-pressure environments. So if you need a job in the next 30–60 days, you cannot build a professional network from scratch fast enough for it to matter. Real relationships take months, and you don’t have months.

So what do you do when you have no connections and you need a job now? You make your cold application so good that it doesn’t need a referral to get noticed. That’s the game I played, and it worked twice.

The math you’re up against

Let’s be honest about what cold applying without connections looks like in 2026. A decent remote role in tech will attract roughly 1,500–2,000+ applicants in the first few days. If you assume maybe 10% of those are genuinely qualified, you’re competing against 150–200 people for a single position. And a good chunk of those 200 have a referral, an internal connection, or a name the recruiter recognizes. (I break down the full math behind these numbers here.)

Your odds as an unknown cold applicant? Honestly, not great on any single application. That’s why you need two things working in tandem: volume and quality. Enough applications that the math works in your favor, and each one good enough to actually compete.

What actually works when you don’t know anyone

Here’s what I learned from getting hired twice without connections. None of this is complicated, but all of it requires effort.

1. Read between the lines of job postings

Most people read a job description and try to match keywords. That’s the wrong approach. What the position really wants in a candidate is almost never written in bold with the exact tech stack and projects they want to put you on. That’s too much information to give to competitors and too specific for candidates to build a profile around.

So they give you hints. A data scientist posting mentions “experience with bots” in passing. What they actually want is someone who has built bot-detection algorithms. An analyst posting mentions “geospatial experience.” They’re looking for specific GIS technologies and libraries, not just someone who knows the word “geospatial.”

When you have no connections, you can’t call someone at the company and ask what the team actually needs. Your resume has to demonstrate that you read between the lines and understood. Ask yourself: what are they really looking for, and what specific experiences are going to convince them of that?

Surface-level keyword matching
“Experience with bots and geospatial data analysis”
This just parrots the posting. It tells the recruiter nothing about what you actually did.
Reading between the lines
“Designed a bot-detection algorithm using session telemetry that reduced fraudulent ad clicks by 22%”
This shows you understood what they meant by “experience with bots” — and that you’ve actually done it.

This is the single biggest differentiator between cold applications that get interviews and cold applications that get ignored. Without a referral pushing your resume to the top of the pile, the resume itself has to do the convincing. And it convinces by being specific to what this job actually needs.

2. Focus on impact, not implementation details

Most people — especially juniors — write resume bullet points about what tools they used rather than what they accomplished. “Used Python to build ML models.” This is about as useful as saying “Used lab notebook to do chemistry research.” The focus is on an implementation detail rather than on the substance.

When you have no connections, your resume is your only advocate. It needs to communicate value, not just activity. What changed because of your work?

Implementation-focused
“Used Python and scikit-learn to build machine learning models for the marketing team”
So you used a programming language to do your job. What happened as a result?
Impact-focused
“Reduced customer acquisition costs by 20% by designing and implementing A/B tests to optimize offer pricing”
Now you’re talking. The implementation details will come out in the interview.

If you want to highlight experience with a specific language or tool, put it in the skills section. Your bullet points should be about what you achieved, not what you used to achieve it.

3. Don’t let formatting kill your chances

This matters more than people think. If an ATS can’t parse your resume, a human will never see it. But the complexity of making an “ATS-friendly” resume is overblown. You don’t need an “ATS specialist.” Avoid columns, avoid symbols and icons, and use a normal resume template. That’s it for the ATS.

The real formatting problem is what happens when you use AI to help with your resume. ChatGPT and Claude will almost certainly still produce ATS-parsable output, but the formatting will look sloppy to a human. Inconsistent spacing, weird bullet characters, lost alignment — the kind of thing that makes a recruiter unconsciously think “this person doesn’t pay attention to details.” (There’s a deeper dive on why ChatGPT destroys your formatting and how to fix it.)

When you have no connections, you can’t afford to give a recruiter any reason to skip you. A messy-looking resume is an easy reason.

4. Find the right volume — the 5–10 minute rule

This is the question at the heart of every job search: should you apply to more jobs, or spend more time on each application?

Both extremes are wrong. Spending 30+ minutes on a single cold application is doing yourself a disservice — the math doesn’t justify that investment when your odds on any individual role are so low. But spending zero time tailoring your resume and just pumping out the same generic application everywhere is equally wasteful. The response rate on untailored resumes is close to zero.

The sweet spot is about 5–10 minutes per cold application. Enough time to read the job description carefully, identify what they’re really looking for, and adjust your resume to speak to that. Not so much time that you’re agonizing over every word. (This is the balance we talk about in do you need to tailor your resume for every job? — short answer: yes, but strategically.)

At that pace, you can do 6–12 quality applications per day. Over a month, that’s 120–250 tailored applications. With the math we discussed above, that range starts to produce results.

5. Don’t sound like AI wrote your resume

This has become a real problem for cold applicants in 2026. Recruiters are flooded with resumes that all sound identical — the same “spearheaded cross-functional collaboration” language, the same perfectly structured bullets, the same suspiciously precise alignment with the job description.

When you have connections, a referral can vouch for you as a real person. When you’re cold applying, your resume is the only evidence that a real human is behind it. If it reads like ChatGPT output, you’re giving the recruiter an easy reason to pass — not because they hate AI, but because it signals you didn’t put in real effort. (I cover the full breakdown of what actually triggers the “this is AI” reaction from recruiters.)

Use AI to help you refine and tighten your language. Don’t use it to generate your resume from scratch. The difference is obvious to anyone who reads resumes for a living.

Why connections matter — and how to close the gap without them

I’m not going to pretend connections don’t matter. They do — a referral gets your resume moved to the top of the stack, an internal advocate means your application gets a closer look, and someone who knows the hiring manager can tell you exactly what the team needs.

But here’s what people get wrong: they think the advantage of a connection is access. It’s not. It’s information. A connected candidate knows what the team actually cares about, so they can tailor their resume precisely. They know the pain points, the projects, the gaps.

You can close most of that gap without knowing anyone at the company. The job description, the company’s engineering blog, their public GitHub repos, their product — it’s all information about what they care about. The candidates who compete with referrals and win are the ones who do the homework that a referral would have handed them for free.

The bottom line

Getting a job without connections in 2026 is harder than getting one with them. That’s just reality, but “harder” is not “impossible” — and I know that because I did it twice.

It comes down to this: when nobody is vouching for you, your resume is doing all the talking. It needs to show that you understood the role, that you’ve done relevant work, that you can communicate your impact clearly, and that a real person — not a language model — is behind it.

You don’t need to know someone at the company. You need to know what they’re looking for — and then show them you’re exactly that, in your own words.

If you want role-specific advice for cold applying, I’ve written deeper dives for tech roles, data roles, and business roles.