I’ve always had a gift for pattern recognition. Some people call it intuition. Some call it strategy. I call it my superpower. I’ve honed it over years of observing and listening. 

And lately, the patterns I’m seeing in tech, SaaS, and AI? They’re very clear to me. And if you’re not paying attention, they’re going to catch you off guard.

Here are three predictions for the next 18 months. This is based on how people are behaving, the media’s coverage of shifts, and the overall state of the world. These predictions are NOT based on what product teams are dreaming up in a vacuum, because they think its cool, or it will differentiate the product.

Prediction 1: The Gen AI Hype Dies Down 

AI is powerful, we get it. But the novelty has worn off. The heavy focus on Gen AI is already shifting, and here are the signs of the times.

Reason 1: We’ve had enough of the unedited, low-effort content

Every feed, every doc, every pitch deck has started to feel the same. You can spot the AI voice from a mile away. It’s flat, filled with jargon, talks in circles, and is often poorly written. Even Linkedin’s algortihm has deprioritized text only posts.

Reason 2: AI-for-the-sake-of-AI is over, and adoption is lagging

Companies rushed to add AI to their products without asking if it helped. Now they’re realizing that just because it sounds good in a board meeting doesn’t mean users are adopting it.

• BCG found that only 26% of companies get meaningful business value from AI initiatives.
• McKinsey reports that most AI projects don’t scale past the pilot phase.
• Comscore shows usage skewing toward basic tasks, and not really supporting deep workflows.

The fantasy was: “Shrink headcount, keep revenue, let the AI do the rest.” ( That was the fantasy, no? Duolingo got backlash for admitting it out loud. ) The reality is GenAI tools still need handholding, and most users aren’t trained for that. Furthermore, most applications are useless. 

For example, I tried using the Alt-text feature in Stensul. What it generated was literal nonsense. I ended up just writing the Alt-text myself. It did not save me time or effort. 

Reason 3: The shiny new thing is now commonplace

AI’s not new anymore. It’s expected. Which means it has to work. And more importantly, it has to disappear into the background. The best AI moving forward won’t be obvious. It’ll be smart enough to show up where you need it and quiet enough to stay out of your way. 

Some orgs are already adding AI features that still have the AI label, but that do great work.

For example, the AI features in ActiveCampaign help you optimize your campaigns. 

Fellow, the note-taking app, takes AI-powered notes for you. It also gives you pre-meeting summaries and automated post-meeting follow-ups. This AI actually supports and streamlines workflows.

I also think Demandbase is getting the AI implementation thing correct. Agentic AI to enhance your workflow, not hinder it. 

Prediction 2: From Acquisition-Obsessed to Customer-Obsessed

We’re watching a big shift, and it’s overdue. The economy is in flux. Given how erratic this White House has been on everything from tariffs to straight-up bombing the people we get our gasoline from. Wallets are closing, and belts are tightening. Those things always have business implications, even in B2B spaces.

Reason 1: Acquisition is getting more expensive

Cost-per-click is up. Ad fatigue is real. Cold outreach is ice cold. ( Especially for those AI-generated cold email templates) If you’re spending more and getting less, maybe it’s time to look inward.

Reason 2: Retention is the multiplier

It doesn’t matter how many customers you bring in if you can’t keep them. And most churn doesn’t happen because your product is bad. It happens because your onboarding, education, and support fall short. You want stickiness? Empower your users.

Reason 3: The industry is already moving

Anecdotally, I’m seeing a shift in hiring too. My LinkedIn job alerts showed a 37% spike in open Customer Marketing roles in just one week. That’s not a definitive report, but it is a signal.

Why? Because smart companies know that loyalty, referrals, and expansion don’t come from luck. They come from listening and delivering. The orgs that prioritize post-sale relationships will win. The ones that don’t? They’ll be stuck re-acquiring their own churn.

Prediction 3: Workflow Optimization Will Outpace Gen AI

People don’t want more tools. They want fewer steps. People want their actual time back, for the projects that move the needle. Increasingly, orgs are exploring partnerships via integration to support smoother workflows that their customers are actually asking for. 

Reason 1: Everyone wants to move faster

We’re drowning in admin, tabs and tools that don’t talk to each other. The most loved tools in 2025 will be the ones that shrink the to-do list. 

Reason 2: Deep integrations will win

Figma is in Notion; Canva is in HeyOrca; and Google Drive is integrated with Slack. That’s what we’re going to see more of. Not chatbot sidebars that do nothing helpful. At this point, the chatbots feel like Clippy revived. 

If I can update my status in Asana just by hitting “Ready to Review” in HeyOrca? That’s real innovation. If I have to copy, paste, switch tabs, and go through a buggy integration? You’ve lost me and probably everyone else too.

Reason 3: People are literally screaming for this

Scroll any product feedback thread. People aren’t asking for more AI-generated content. They’re asking:

“Can you just automate the boring stuff for me?” “Why do I still need three tabs open to do this one task?”

They don’t want Gen AI. They want agentic AI, automations that handle the handoffs so they can focus on the real work.

Don’t miss the boat

The rapid growth of tools like ChatGPT showed us what’s possible. It was exciting and possibly world changing, with lots more downside than upside. It has been shown to negatively impact the human brain, and drive skill loss in heavy users according to MIT.

We’re moving into a new era now. One where hype gives way to utility. Where the winners won’t be the flashiest tools, but the most helpful ones.

If your product can help people think less about their workflow and more about their work? That’s where you’ll win. 

The TLDR of it all, Listen to your customers, and pay attention to how they’re actually using your product.”

Now, is it pattern recognition or psychic powers? You tell me.