Just Add AI (And Other Lies We’ve Told Ourselves)

The Problem: “Just Add AI” Isn’t a Strategy

There is a phrase showing up in strategy meetings, vendor demos, and executive planning sessions that sounds decisive enough to pass for strategy – even when it isn’t:

“We just need to add AI.”

It shows up at predictable moments. Marketing wants more output without more headcount. Sales wants better follow-up and clearer priorities. Leadership wants reporting that actually says something. Service teams want fewer repetitive questions without slowing down response time.

Somewhere in the middle of all that, AI starts to sound like a clean answer to problems that are anything but clean.

The appeal is real. In a remarkably short period of time, AI has moved from novelty to expectation. A prompt can now generate a draft landing page, summarize research, produce a proposal outline, suggest outreach language, or help build a functioning prototype before most teams have finished discussing who owns the next step. That kind of speed is useful, and it’s changing how work gets done.

What it doesn’t do is replace clarity.

The Reality: AI Magnifies What’s Already Broken

That is where most businesses collide with reality. AI is often introduced as though speed alone will compensate for unclear systems, uneven handoffs, weak data discipline, or messaging that was never particularly sharp to begin with. In practice, AI magnifies what’s already there. If your structure is strong, AI creates lift. If it’s messy, AI exposes it – fast. If your CRM is a digital junk drawer, AI just makes the drawer louder.

McKinsey & Company reported in its 2025 State of AI research that 78 percent of organizations now use AI in at least one business function, yet only a much smaller share describe those efforts as materially affecting enterprise-wide outcomes.

Data Point: 78% of organizations now use AI in at least one business function, but far fewer say it is materially improving enterprise-wide outcomes.

The takeaway is simple: experimentation is everywhere. Real impact isn’t.

The gap isn’t about technology. More often, it is operational readiness, because AI only performs as well as the environment it enters.

A weak sales process doesn’t become stronger because emails are faster. A crowded CRM doesn’t become strategic because summaries now appear automatically. Unclear positioning doesn’t become differentiated because AI can produce five versions of the same message instead of one.

This is exactly why the newest generation of AI tools deserves both enthusiasm and restraint. The tools themselves are impressive. The bigger question is whether the business using them is ready to turn speed into actual advantage.

The Shift: From Tools to Systems

Tools like Lovable have gained attention because they allow teams to describe a digital product in plain language and quickly generate a functioning interface. Replit has accelerated that same idea by making it easier to build and deploy lightweight applications, internal tools, and working prototypes inside a live development environment. For companies used to waiting on development queues, this can feel transformational because ideas move visibly faster. 

AI Tools Worth Watching

• Lovable: Interface generation from plain-language prompts
• Replit: Fast prototyping and deployment
• Cursor: AI inside the development workflows
• Perplexity: Research with source transparency
• Synthesia: Scalable video creation
• Gamma: Executive-ready presentations built quickly
• What problem are we actually solving? AI without purpose creates noise.
• Is our data trustworthy? Bad inputs weaken every output.
• Where should AI assist – and where should humans decide? Not every decision should be automated.
• Does this strengthen a system or create another disconnected layer? Disconnected tools create fragmentation.
• Are we using AI to create clarity, or avoid it? Speed without clarity rarely converts.

What these tools all share is that they reduce production friction. What they do not do is replace judgment.

That readiness matters even more when AI moves beyond isolated productivity tools and begins operating inside systems that hold customer relationships, pipeline movement, service history, and decision-making context. That is where HubSpot is becoming especially interesting.

Where It Gets Real: AI Inside the CRM

Much of the early AI conversation focused heavily on marketing because marketing produced the easiest demonstrations. A prompt could generate campaign language instantly, suggest social copy, or help structure a landing page before the meeting was over.

HubSpot’s Breeze framework reflects that shift. Rather than treating AI as a separate utility layered on top of the platform, HubSpot is building assistants and agents that work within the customer platform itself, where CRM records, timeline activity, deal stages, service conversations, and internal knowledge already exist.

A writing tool drafts a message. AI inside your CRM understands the context behind it – whether that message relates to an open opportunity, a stalled account, a recent service issue, or an active buying conversation that has already involved multiple stakeholders.

Nowhere does that become more practical than on the sales side, where small delays, inconsistent follow-up, and weak visibility often create larger revenue consequences than teams realize.

Here’s what that looks like in practice:

Rather than simply helping draft outbound language, it is designed to monitor CRM signals, engagement patterns, and prior contact activity so that outreach begins with context rather than guesswork.

The practical gain is subtle but important. A rep no longer has to manually rebuild opportunity priority each morning.

A recent conversation with a senior leader inside a large financial institution captured this shift well.

“For the first time, our team is not guessing where momentum is coming from. We can see where conversations begin, where they stall, and what is actually moving people toward conversion.”

What stood out wasn’t just visibility. It was how quickly leadership began trusting the process once sales activity, follow-up discipline, and lead movement became visible in one place. AI then became useful because it had clean signals to work from, not because it replaced the underlying sales strategy.

HubSpot has also expanded AI directly into the sales workspace, allowing sellers to navigate a summarized communication history, suggested next steps, and deal context without manually reconstructing every conversation.

This is why AI inside your CRM matters more than AI outside it. The system already holds the memory of the relationship.

Even the Breeze Customer Agent, often discussed through a service lens, has growing relevance for sales because buying conversations do not always begin neatly during business hours.

HubSpot’s Knowledge Base Agent adds another layer that often matters more over time than many teams initially expect.

The broader point is not simply that HubSpot now includes AI. Many platforms can claim that. What makes this worth watching is that HubSpot is embedding AI into the processes businesses already use to manage relationships, make decisions, and drive revenue.

Before You “Just Add AI”

Five Things to Think About Before You Just Add AI

The organizations seeing the strongest results right now are rarely the ones chasing every new tool. They are the ones disciplined enough to understand where AI belongs, where human judgment still matters most, and how both should work together without creating noise.

If AI is exposing cracks in your systems rather than improving them, that’s not a platform problem – it’s a structural problem. GROWL helps you clean up the data, align your systems, and operationalize your CRM so AI actually drives results. Not more noise.