I Cut My SaaS Tool Stack From 23 to 9 Tools (And Revenue Didn't Drop)

Two years ago I was running 3 SaaS products and paying for 23 different tools.

Notion, Slack, Stripe, AWS, Vercel, Mixpanel, Intercom, Linear, Figma, Loom, Calendly, Zapier, Firebase, Postmark, CloudFlare, MongoDB, Algolia, Segment, LogRocket, Sentry, GitHub, Zoom, and whatever else I convinced myself I "needed."

Monthly bill: $847.

Today I run 10 SaaS products. My tool bill? $214/month.

Revenue didn't drop. Speed didn't drop. Quality didn't drop.

Here's what changed.

The Wake-Up Call: $10K/Year on Tools I Opened Twice

Last January I did a brutal audit.

I went through every subscription and asked:

1. Did I use this in the last 30 days?
2. If I cancelled it right now, would I notice within a week?
3. Is there a free/cheaper alternative that does 90% of what I need?

Turns out:

• 9 tools I hadn't opened in 2+ months
• 7 tools I could replace with free alternatives
• 4 tools doing the exact same thing (because I forgot I already had one)

I was paying for Intercom ($150/month) and using it to send 6 emails a month. I could've used Postmark ($10/month) for that.

I was paying for Mixpanel ($89/month) when Google Analytics + Plausible (free tier) gave me everything I actually looked at.

I was paying for Linear ($8/seat) when I'm a solo founder. GitHub Issues is free.

Classic.

The 9 Tools I Actually Use (And Why)

Here's my entire stack now. Everything else is gone.

1. Supabase ($25/month)

Replaces: Firebase, MongoDB, custom backends, auth services.

Why I kept it: Postgres + auth + realtime + storage in one place. I build SaaS apps — this is my entire backend.

I used to run separate services for database, auth, and file storage. Now it's one bill, one dashboard, one less thing to think about.

If you're building SaaS and not using Supabase, you're wasting time stitching together 4 tools.

2. Vercel ($20/month)

Replaces: AWS, Netlify, custom servers, DevOps stress.

Why I kept it: Deploy Next.js apps in 30 seconds. Zero config. Scales automatically. I don't want to think about infrastructure.

I tried self-hosting on DigitalOcean to "save money." Spent 6 hours fixing a deployment issue. Vercel costs $20/month and deploys work every single time.

That's $20 well spent.

3. Cloudflare ($0 - $20/month)

Replaces: AWS CloudFront, Fastly, DDoS protection, DNS management.

Why I kept it: Free tier handles 90% of what I need. CDN, DNS, DDoS protection, Workers for edge functions.

I moved all my domains to Cloudflare. The analytics alone would cost $50/month elsewhere.

Plus Cloudflare Workers let me run serverless functions without dealing with AWS Lambda configs. Clean.

4. Notion ($10/month)

Replaces: Evernote, Google Docs, Trello, Airtable, a dozen note apps I tried and abandoned.

Why I kept it: It's my second brain. Product specs, user research, todo lists, knowledge base, everything.

I tried ditching it for free alternatives (Obsidian, Logseq). Came back in a week. Notion's databases + relational links are too good for managing multiple products.

Worth every penny.

5. Tally ($0 - $29/month)

Replaces: Typeform, Google Forms, SurveyMonkey.

Why I kept it: Free tier is insane. Unlimited forms, unlimited responses, integrations, and it doesn't look like garbage.

I use it for:

• Waitlist signups
• Customer feedback surveys
• Feature request forms
• Early access applications

Typeform wanted $35/month for 100 responses. Tally gives me unlimited for free. Easy choice.

Try Tally here

6. Plausible Analytics ($9/month)

Replaces: Google Analytics, Mixpanel, Amplitude.

Why I kept it: Privacy-first, lightweight, actually readable.

I don't need funnel analysis for 14 different user segments. I need to know:

• How many people visited
• Where they came from
• What pages they viewed

Plausible does that in a 3-second glance. No cookie banners. No GDPR nightmares. No 47-step setup.

Mixpanel was overkill. This is perfect.

7. GitHub ($4/month)

Replaces: GitLab, Bitbucket, self-hosted Git, project management tools.

Why I kept it: Where my code lives. Issues, PRs, Actions for CI/CD. It's the default for a reason.

I tried Linear for task management. Realized GitHub Issues + Projects does 90% of what I need. Cancelled Linear. Saved $96/year.

8. Stripe ($0 + fees)

Replaces: Nothing. This is non-negotiable.

Why I kept it: You need a way to get paid. Stripe is the best at this.

I looked at Lemon Squeezy, Paddle, PayPal. Stripe's API is still the gold standard. Fees are the same everywhere anyway.

Stripe Billing + Checkout = my entire payment infrastructure.

9. Cursor ($20/month)

Replaces: VS Code, GitHub Copilot, ChatGPT for coding.

Why I kept it: AI-powered code editor that actually works. Writes 60% of my code. Catches bugs before I run the app.

This is the only tool on this list I'd pay 10x for.

I was skeptical of AI coding tools. Then I used Cursor for a week and shipped a feature that would've taken 3 days in 6 hours.

It's not perfect. But it's fast. And speed matters when you're solo.

If you write code and you're not using Cursor, you're working 3x harder than you need to.

What I Cut (And Don't Miss)

❌ Intercom → Switched to plain email (Postmark)

Saved: $140/month

I thought I needed a fancy live chat widget. Turns out users just want to email you.

❌ Mixpanel → Switched to Plausible + spreadsheets

Saved: $89/month

I was tracking 47 events and looking at 2 of them. Overkill.

❌ Linear → Switched to GitHub Issues

Saved: $8/month

Linear is beautiful. But I'm one person. I don't need Gantt charts.

❌ Figma → Switched to Excalidraw + pen & paper

Saved: $12/month

I'm not a designer. I was using Figma to draw rectangles. Excalidraw is free and does the same thing.

❌ Loom → Switched to native screen recording

Saved: $8/month

Mac has built-in screen recording. So does Windows. I was paying for HD uploads I never used.

❌ Calendly → Switched to Cal.com (self-hosted, free)

Saved: $10/month

Cal.com is open-source Calendly. I self-host it on Vercel (already paying for that). Zero extra cost.

❌ Zapier → Switched to n8n (self-hosted) + Supabase functions

Saved: $50/month

Zapier is great until you hit 100 tasks/month and the bill jumps to $50. I moved automations to Supabase Edge Functions. Free.

❌ MongoDB Atlas → Switched to Supabase (Postgres)

Saved: $60/month

Postgres > Mongo for 90% of use cases anyway. Plus Supabase handles backups, scaling, everything.

❌ Algolia → Switched to Postgres full-text search

Saved: $50/month

Algolia is insanely fast. But Postgres full-text search is fast enough for my scale (< 100K records). Saved $600/year.

❌ Segment → Switched to direct integrations

Saved: $120/month

I was using Segment to send data to 2 places. Just integrated those 2 places directly. No middleman needed.

❌ LogRocket → Switched to Sentry (free tier)

Saved: $99/month

Session replay is cool. Error tracking is essential. Sentry's free tier gives me the essential part.

❌ AWS (for side projects) → Switched to Cloudflare Workers + Supabase

Saved: $40/month

AWS bills are impossible to predict. I'd get hit with random $60 charges for things I forgot were running. Moved everything to flat-rate tools.

Total saved: $633/month = $7,596/year

The Real Cost of Tool Bloat

It's not just the money.

Every tool you add:

• Is another login to remember
• Is another dashboard to check
• Is another integration to maintain
• Is another thing that can break

I used to spend 30 minutes a day just switching between tools.

Now I have 9 tabs open max. Everything lives in Notion, GitHub, or Supabase.

Focus returned.

The "Do I Actually Need This?" Test

Before adding any new tool, I ask:

1. Can I do this with a tool I already have?
2. Will this save me more than 2 hours a month?
3. Is this solving a real problem or a hypothetical one?

If the answer to #1 is yes, I don't add it.

If the answer to #2 is no, I don't add it.

If the answer to #3 is "hypothetical," I don't add it.

Most tools fail this test.

What About Enterprise Tools?

People ask: "What if I need to scale?"

Here's the truth: You'll know when you need to scale.

When Plausible can't handle your traffic, you'll see it break. Then you upgrade.

When Supabase's free tier isn't enough, you'll hit the limit. Then you pay more.

Don't pre-optimize for problems you don't have.

I wasted $200/month on "enterprise-grade" tools when I had 50 users.

Stupid.

The Tools I'd Never Cut (Even at Gunpoint)

• Supabase: My entire backend. Non-negotiable.
• Vercel: Deployment sanity. Worth every penny.
• Cursor: 10x productivity boost. Would pay $100/month.
• Stripe: How I get paid. Obvious.

Everything else is replaceable.

My Current Monthly Bill Breakdown

Real average: $214/month (some months Supabase/Vercel spike if traffic is high).

Down from $847.

What I Learned

1. Free tiers are underrated.

Cloudflare's free tier handles more than most paid CDNs. Tally's free tier beats paid form builders. Use them.

2. Postgres can replace 4 tools.

Full-text search, queues, cron jobs, analytics — Postgres does it all. Stop adding tools.

3. "Best in class" is usually overkill.

You don't need the best analytics tool. You need one that works.

You don't need the best form builder. You need one that doesn't suck.

Good enough beats perfect-and-expensive.

4. Tool consolidation = mental clarity.

Fewer tools = fewer decisions = more time building.

I switched from 23 tools to 9 and felt my stress drop by 40%.

5. Pay for speed, not features.

Cursor costs $20/month. It saves me 10 hours/month. That's $2/hour for speed.

Intercom cost $150/month. It saved me 0 hours/month. That's $150 for nothing.

Pay for what makes you faster. Cut everything else.

How to Cut Your Own Tool Stack

Step 1: List every tool you pay for.

Step 2: For each tool, ask: "If this disappeared tomorrow, would I notice within 3 days?"

Step 3: If the answer is no, cancel it. Today.

Step 4: For tools you keep, ask: "Is there a free or cheaper alternative that does 80% of this?"

Step 5: Try the alternative for 2 weeks. If it works, switch.

Step 6: Repeat every 6 months.

I do this audit twice a year. Always find at least 2 tools to cut.

Final Thought

More tools don't make you more productive.

They make you more distracted.

The best tool stack is the smallest one that gets the job done.

Mine went from 23 to 9. Saved $7,600/year. Shipped faster.

Yours can too.

---

I'm Kapil — solo SaaS founder, serial tool-cutter, recovering subscription addict. I build products with 9 tools and refuse to add a 10th. Follow me on Twitter for more honest takes on bootstrapping.