🏠 Home | πŸš€ Posts | πŸ“ Guides

#002 jobGen - The Grand Romantic Gesture of a Hacker

My spouse had a pain point! So I built her SaaS to solve it

jobGen - generate job descriptions with ease

πŸ€” The Problem

My spouse's startup is growing like crazy. And as an HR manager, she works around the clock to hire the best of the best.

As we both started working from home and my wife became my co-worker, I've noticed that she writes a lot of job descriptions.

So, I thought, maybe I can build a product that will help her write job descriptions?

πŸ” Research

The first step was to research. My solution can have two benefits: (i) save time (ii) write better job descriptions. So, what makes a job description great?

A job description for recruitment is what a landing page for SaaS. It's the first encounter a candidate has with the company.

Similar to a landing page, the job description needs to convert. Get the candidate to submit their resume. And maybe more important, convert the right candidates. Candidates that will be a good fit for the company and the position.

After doing some research, these are the three top things to improve your job descriptions:

  1. Summary: The first time a candidate sees your job description, they scan it for 5 seconds. Their main questions - what's the company and what's the role? Include a short summary at the top, answering these questions. Your conversion rate will sky-rocket.
  2. Clear Sections: Once you pass the quick scan test, the candidate dives deeper. After the summary, your posting should have two clear sections - one about the role, and the other is about qualifications (optional - benefits section). Don't write about the company beyond the summary. If the candidate wants to learn more, they'll go to the company website.
  3. Use accurate bullets: Use bullets, it's easier to skim. And, make sure to only include the most critical bullets. If you need 3 years of experience, include 3 years - not 1 and not 5. Remember, your goal is to convert the best fitted candidates. Too generic bullets will make skillful candidates pass. Too many demands, might get great candidates to move on, thinking they don't have what it takes for the role.

πŸ’‘ Solution

My solution is simple. Help users by generating a role summary, suggesting bullets for the role description & qualifications, and create an easy to scan job description that is divided to three clear sections:

  • Summary
  • About the role
  • Qualifications

All within a few minutes and following the best practices above.

To build this, I took 750 job titles. For each one, I fetched a summary and then scanned hundreds of real job descriptions to scrape bullets and classify them to "about the role" and "qualifications".

Once I have the data, all I needed is to develop a web app where users can input information about a role and a job description with suggested bullets is generated.

βœ… Validation

I decided to do a quick validation.

I created a landing page with email collection, and published it on Reddit.

13 up-votes and 40 email sign-ups was enough for me to call it "validated". I thought to myself, if in 30 minutes of work I could get 40 emails, I should build it (⚠️WRONG⚠️).

I also mailed my 40 users asking for an interview. 2 replied and scheduled.

πŸš€ Launch

So I built my app, named it jobGen, revamped my landing page and I was ready to go. I targeted Growth stage HR managers (e.g. my wife), and found a few subreddits, Linkedin and Facebook groups that are relevant.

I debated with myself about whether I'd like to charge money from day 1. I decided to keep it free.

Now all is left is to push the button

AND... a flop.

Launch statistics 60 visitors


  • This entire launch only brought 60 visitors to the landing page
  • Out of the 60, 20 signed up (decent conversion rate)
  • Only 31% of my email list even opened my email (see email below)
  • 100% of mail opens converted to users

Launch email to waiting list


So all in all, given that I published my app in several places and I already had 40 emails, the results were not good.

πŸ“‰ Post-launch

My wife found my product super useful. She actually uses it frequently and loves it. I guess that's the most important thing.

Beyond that, the results post launch were even worse than the launch itself. None of my non-wife users use the product regularly. They all churned after 1-30 days.

I tried to source a few interviews to see why users are not as engaged as I hoped. No one replied, even though I followed up a few times.

Then I said "Fine. Maybe an interview is too much. I'll create a 2-min survey". Even to that only 2 users replied

πŸ“§ Email campaign

As a last effort, I decided to create an email marketing campaign. I had 20 unengaged users + 30 people who provided their email but never signed up. I wanted to see if I can get a few active users out of the 50 emails.

Over the course of 10 days, I sent 5 emails:

  1. The story behind jobGen (my wife's pain-point)
  2. What makes a great job description and how jobGen helps create them
  3. Call to action to try jobGen for free and asking users who are not interested to opt-out so their place can go to next in line (my desperate attempt to create FOMO)
  4. Iterating on how jobGen can help my users
  5. Thanking my users and asking for feedback about jobGen (desperate attempt to create social proof)

No success!

None of my non-users decided to sign up. For my unengaged users, some tried jobGen again, but no one became regularly active. Also, I had low email open rates (20%).

πŸ“ Conclusions

So overall I'm left with one, very important, active user (my wife).

Play on easy-mode

This relates to another learning I had from my previous post. There, I launched a Netflix recommendation app and got 10k visitors in one day. I was playing on easy mode. It's a pandemic and streaming services are the hot thing. I didn't need much to get visitors to my website. Here, I was playing on hard-mode. I chose a target audience that I'm not familiar with, I had no connection to (except my wife), HR managers are a target of many SaaS so most groups forbid promotions, etc...

Target audience

The main pit-fall here is that my users are not my target audience. I wanted to target HR managers in growth stage companies. They hire fast, in high volume, and often for new roles. But, most of my users work at recruiting firms (checked by the email used for sign-up).

Email list

Email list as validation should contain many emails. 5X the number of users you hope to acquire at launch. Or try to search other signals in addition to the mailing list. For example, you can test the engagement of your mailing list by sending them a request for survey/interview before starting building.

I'm left more with questions then answers:

  • Is 20%-30% email open rate normal? How can I increase it in future projects? Can I avoid landing in promotions tab?
  • How could I break the negative cycle of - my users are not engaged so they don't want to help with interviews. But, I can't improve my product and get users engaged without interviewing them.
  • Am I targeting the right audience, and how can I reach them?
  • Should I have charged for jobGen? That way my audience would probably be smaller, but also more engaged. Or maybe I would have been left with 0 users, 0 learnings? (and more development time to integrate Stripe)

Let me know if you have any insights to share. You can DM me on Twitter or reply to my tweet

⏭️ Next steps

I have a few options. Still debating which one I should choose. I welcome any suggestions.

  1. Cut my losses: Conclude that my app doesn't solve a real problem, and be happy that at least my wife loves it
  2. YOLO: Broadcast my app as wide as possible, hoping to stumble upon some users who like it. Launch on PH, IH, etc, etc.
  3. SEO: SEO can be a good match here. Many people search "Job description for role YYY example". But given that SEO takes >6 months, I'm not sure I should spend so much effort with so little validation
  4. Targeting: Do more research and find a way to market to HR Managers. That would be ideal, but till date, I didn't find a good channel. Most subreddits/groups ban promotions and the ones that allow it, I already included in my initial launch.

Thanks for reading my post! if you liked it, share the πŸ’– in the tweet below and/or let me know your thoughts.