SaaS Truths No One Talks About: PLG, Churn, and Getting Users That Stick (Part 1)

This content originally appeared on DEV Community and was authored by Mukul Sharma

If you’re building SaaS right now, you’re in the heat. AI’s reshaping the game daily, funding’s drying up, and everyone’s still chasing hockey-stick growth – but with way less room to fail.

What’s working now isn’t flashy tactics or Twitter threads. It’s what seasoned founders are doing behind the scenes – stuff that’s messy, thoughtful, and battle-tested.

I pulled insights from a podcast stacked with founders like David Kelly (AppSumo Originals), Nad Lazaric (SEO powerhouse), and others who’ve scaled to multi-million ARR. Their lessons? No fluff. Just what actually works when you’re deep in the trenches.

This is Part 1 of a two-part series breaking down 7 sharp, honest SaaS lessons. Let’s dig into the first four:

1. Product-Led Growth? Great If It Fits-Disaster If You Force It

PLG sounds like a dream: build something great, let users sell it for you. No sales team, no hand-holding.

But here’s the reality:

Only one founder in dozens of interviews made PLG truly work.
David Kelly, CEO of AppSumo Originals, hit $13M ARR using PLG – but only because he picked markets where the product naturally spreads.

His formula:

“We target spaces where the product inherently communicates with potential users-co-workers, clients, whoever.” – David Kelly

Why PLG Fails for Most:

• Your product isn’t naturally visible to others.
• You assume “build it and they will come” still works.
• You never test virality early.

Takeaway:

Audit your category:

• Does using your product expose it to new people?
• If not, layer in outbound sales, content, or partnerships – early.

2. AI’s Wrecking Content Marketing – But Reddit’s Your New SEO Lifeline

Content still converts in SaaS – but AI’s flooding the web with low-quality junk. Google’s fighting back. So are users.

Nad Lazaric, an SEO pro, ran tests on AI content and found:

“A lot of AI-generated posts are pretty low quality. Google won’t penalize AI – it penalizes low value.”

What’s changed:

• Raw ChatGPT blogs? Maybe they rank for a bit.
• Long-term? You’ll get buried unless you edit heavily and add real insights.

Meanwhile, users are adapting:

“People are searching ‘best X tools Reddit’ just to avoid SEO spam.” – Lioren, ex-Kissmetrics

Reddit’s become the go-to for authentic answers – and Google knows it.

Why Reddit Works in 2025:

• Signals authenticity + user-generated trust
• Subreddits rank high on Google
• Easy to engage, seed, and even moderate to guide visibility

Quick Comparison: SEO in 2020 vs. 2025

Takeaway:

• Stop shipping unedited AI posts.
• Build or join relevant subreddits.
• Comment, share, ask, mod – and watch your SEO climb without paying Google a cent.

3. Churn Control? It Starts Way Before Signup

Churn’s the quiet killer. Doesn’t matter how many users you add if you’re losing them every month.

Aaron Krall (AgentMethods) and Tim Cool (Smart Church Solutions) flipped the script:

“We don’t let just anyone sign up. If you’re not a fit, we don’t want you.” – Aaron Krall

What They Stopped Doing:

• Self-serve free trials
• Letting brand-new businesses onboard
• Chasing every inbound lead

What They Do Instead:

• Mandatory demos or onboarding
• Use data to predict churn risk before signup
• Charge upfront (yes, even for onboarding)

Example:
Aaron found insurance agents <12 months into business churned at ~50% – many left the industry in 13 weeks. So he stopped accepting them. Churn dropped. Support costs dropped. Growth improved.

“It’s deliberate – pricing, training, selectivity.” – Tim Cool

Result?

• 4–5 year customer lifetimes
• Net negative churn
• Fewer headaches

Common Churn Triggers to Track:

Takeaway:

• Use your data: who churns and why?
• Don’t be afraid to say no early.
• High-quality, sticky users > high signup volume.

4. Human Touch Wins: Ditch the Bots, Pick Up the Phone (or Plane)

Founders want scale. But users want connection. And in the AI age, human interaction is a competitive advantage.

Nick McHenry (OneShop) burned $1.2M on ads and automations. Nothing stuck – until he started meeting customers in person.

“That separated us from competitors.” – Nick McHenry

Mehdi Ali (HelloWash) used AI agents… then pivoted to cold calls – himself.

“No pitch. Just short, friendly calls to book a demo. Fits my schedule. Gets results.”

And Krall again:

“Dev founders want to self-serve. Insurance agents? They want a call. Match the style to the customer.”

3 Low-Key Human Plays That Still Work:

1. Cold calls during your downtime → 2–3 calls a day = 10+ demos a week
2. Customer visits / meetups → Build trust + gather real feedback
3. Paid onboarding with actual humans → Higher activation, better retention

Takeaway:

Human > automation, especially in traditional industries. Don’t hide behind bots. For introverts? Start small – 1 conversation a day.

Final Thoughts

That’s the first four truth bombs from SaaS founders in the arena. None of these are “growth hacks” – they’re structural choices that fix broken assumptions.

If you’re building right now, try even one of these, and you’ll feel the shift.

Coming up in Part 2:

• How to unlock powerful secondary revenue streams
• Mastering cold outreach that actually works
• Why podcasts remain one of the best growth hacks today

Which of these four hit hardest for you? Share it in the comments – and if you’ve learned something similar while building your own product, I’d love to hear it.

This content originally appeared on DEV Community and was authored by Mukul Sharma