Less is more, even in a broken job market

The spray and pray phase

If you are wondering how many jobs you should apply to, the honest answer is probably fewer than you think. More applications only help if the roles actually fit and your CV can speak clearly to them.

For a long time my job searching looked like this. See a title that sort of matched, skim the JD, tweak the CV a bit, hit apply. Repeat until I was too tired to think.

I told myself it was a numbers game. More applications, more chances. In reality I was just adding noise to my own search and getting ghosted at scale. The CIPD’s Resourcing and Talent Planning report has been saying for years that employers are drowning in applications and struggling to filter, and LinkedIn’s own Easy Apply documentation is a reminder of how little friction there now is between seeing a role and firing off another generic application. I was part of that problem.

If you are spread across job-search websites, juggling other job boards, and hitting quick-apply buttons because the job listing looks vaguely plausible, it is very easy to confuse activity with progress.

At some point I realised I was not being strategic. I was coping.

How many jobs should you apply to? Usually fewer than you think

Sending 200 half-relevant applications does not make you look keen. It just means 200 versions of you with a slightly off CV floating through applicant tracking systems that were never going to shortlist you.

What actually moves the needle is fewer, better targeted applications where:

• The role genuinely fits your experience.
• Your CV speaks to what the JD is really asking.
• You have a clear reason for applying beyond “it was there”.

That is the real quality over quantity job applications point. The goal is not to be passive. It is to apply to fewer jobs with a much higher hit rate.

That sounds obvious. The hard part is being honest with yourself about which jobs those are.

The big name trap

Here is where my brain betrays me. A familiar logo shows up and suddenly I am ignoring the fact that the role wants five years of experience in something I have barely touched. I convince myself I can “grow into it” because the company looks good on a CV.

I have wasted hours on applications like this. Nice rejection email, if I got one at all.

The opposite trap: the underqualified illusion

The flip side is just as bad. A JD lists a stack of tools I have not used, phrased in language I do not recognise, and I scroll past. Then weeks later I realise the role was essentially what I do every day, just described in someone else’s vocabulary.

Technical hiring is full of this. One company’s “platform engineer” is another’s “DevOps lead” is another’s “SRE”. The World Economic Forum’s Future of Jobs Report 2025 points to the same broader shift: skills are changing quickly, and employers increasingly have to look past static titles and old credential shortcuts. If you are filtering purely on familiar terms, you are missing roles that actually suit you. The Future of Jobs Report angle matters here because the labels keep moving faster than most job seekers expect.

Why I started scoring jobs before I applied

This is where an honest second opinion helps. I now run roles I am considering through an AI scoring step before I spend any real time on them.

What I want from it is simple. Tell me, based on my actual profile, how well this role matches. Show me where I am strong. Show me where I am weak. Do not flatter me. Do not dismiss me because I phrased something differently from the JD.

That is the part people miss when they talk about scoring jobs. The scoring logic matters more than the shiny output. I do not want a random match percentage floating in the air. I want AI scoring that reads the job description, the job post, the job title, and my actual profile with some respect for context.

The useful thing about AI here is that it does not care about the logo. It does not get excited about a famous employer and it does not get intimidated by a dense JD. It just compares what you have done to what the role is asking for.

A few things changed for me once I started doing this:

• I stopped applying to shiny roles that scored badly once I looked past the brand.
• I started applying to roles I would previously have skipped, because the scoring showed my experience mapped well even when the vocabulary did not.
• I spent more time on fewer applications, and the response rate went up. That is exactly the kind of pattern a proper search performance view should make obvious.

That includes the occasional dream role. Sometimes the dream role is actually a poor fit once you examine the requirements properly. Sometimes it is much more realistic than your own insecurity would have you believe.

What good AI scoring actually looks like

Not all of it is useful. A generic “you are a 72 percent match” number means nothing on its own. The deeper scoring walkthrough is really in My job search experiment with OpenClaw, now turning into a full scale SaaS idea. The short version is that I want:

• A breakdown of which requirements I clearly meet.
• Which ones are partial, and why.
• Which ones I do not meet at all, and whether they are dealbreakers or nice to haves.
• Suggestions for how to frame my existing experience against the JD without making things up.

That last point matters. The goal is not to invent skills. It is to notice the ones I already have but have been describing in the wrong language. Good AI-powered scoring should also notice keywords from the job without turning the answer into AI-generated mush.

How I use AI for this, and why I built OutRung

I started out with a pretty rough setup. Prompts I had cobbled together, bits of my CV pasted into chat windows, scoring logic I was tweaking every other day. My own prototype was wonky, hard to use, and far too technical for anyone who was not me. But even in that messy state I could see it was a diamond in the rough. The signal it gave me was genuinely better than anything I got from scrolling job boards or trusting my own gut.

That is why I ended up building OutRung. Same underlying idea, but a more polished experience so I am not fighting my own tools every time I want to evaluate a role.

The flow is straightforward now. I keep one master profile with my real career history, skills, and achievements. When a role comes up, I score it against that profile before I commit to applying. The scoring gives me the unflattering version of the truth. Sometimes it tells me the dream role is not actually a fit and saves me an afternoon. Sometimes it tells me a role I almost skipped is a strong match and nudges me to look again. Either way I am making decisions with more signal and less ego.

For the roles that do pass the bar, I use the same profile to generate a tailored CV for that specific JD, sense-check it in the ATS score checker, and keep the process visible in the job application tracker. One place, one source of truth, far less admin.

Practical tips if you want to try this

• Write down what you actually want from your next role before you look at any jobs. Seniority, domain, tech, work pattern, salary floor. Use it as a filter.
• Build a proper master profile somewhere. Not a CV, a full record of what you have done. You will reuse it constantly.
• Before applying, score the role against your profile honestly. If it is a weak match and you cannot explain why you still want it, skip.
• Pay attention to roles that score higher than you expected. That is where your blind spots live.
• Only tailor a CV once you have decided the role is worth it. Do not tailor first and rationalise later.

That filter gets easier to keep honest when you can also see salary visibility, role freshness, and repost noise over time, which is the point of the planned Market Activity Tracker.

If you want to include AI in that process, do it where it has clear answer options and real-world constraints. Use it to compare your skills and experiences against the role. Use it to tailor your resume and maybe your cover letters after you have decided the role is worth pursuing. Do not use it as an excuse to spray many jobs with a weak story and hope the volume increases your chances.

Those are basically the frequently asked questions I wish I had answered properly before I wasted so much time on weak-fit applications.

The quiet version of job searching

I apply to fewer jobs now. I spend more time on each one. I get more replies. It is not dramatic and it is not a hack. It is just what happens when you stop treating your career like a lottery ticket and start treating each application like it should actually mean something.

The broken bits of hiring are not going away soon. But I can at least stop adding to the chaos on my own side of it.

That is my answer to how many jobs you should apply to. Enough to stay in motion, but not so many that every application turns into vague wishful thinking.