Clicks Were a Crutch. AI Made Them Obsolete: Rethinking Email Performance in Modern CRM
Written by Christine Harris
For years, marketers have treated clicks as the clearest indicator of interest. A click meant intent. No click meant indifference. Entire KPI frameworks were built around this single binary action.
That system only existed because we did not have anything better.
Today, with Agentic AI and cross-channel intelligence, clicks have become one of the least reliable ways to understand whether an email is doing its job.
The truth is simple:
Email performance is no longer defined only by what happens inside the email. It is defined by how that email influences the customer’s progression across the relationship arc.
Clicks Were a Proxy Because We Lacked Real Signals
Historically, marketers measured what was easy, not what was meaningful.
Clicks were visible, countable, and exportable. They created neat dashboards. They gave strategists something to optimize.
But they were a crude proxy for intent.
A click often reflected curiosity, confusion, device errors, or accidental taps.
Meanwhile, real interest was invisible because we had no way to observe the silent signals.
We optimized for what was measurable, even when it was not the right measurement.
Agentic AI Changed the Rules Completely
Modern CRM platforms no longer rely on explicit actions to interpret engagement.
AI can evaluate a customer’s level of interest using dozens of micro-signals that were previously impossible to measure, such as:
Scroll depth and reading patterns
Time spent on specific modules
The recurrence of certain content affinities
Cross-channel activity that correlates with email exposure
Long-term progression patterns instead of single-session behavior
In other words: the customer does not need to click for the system to know what they care about.
A click tells you what someone did.
AI tells you what someone meant.
This shift fundamentally redefines how CRM success should be measured.
Customer State Changes Are Now the Only KPI That Matters
Most brands still categorize emails by format. Newsletter. Offer. Journey step. Trigger. Honestly, this is understandable because thats how most CRM programs are built, but customers do not experience any of these categories. They experience a series of nudges that shape their readiness, and now we have a way to track and measure that.
The real performance question is no longer, “Did they click”
It is, “Did this email move the customer to a more valuable state”
Examples of state transitions:
Unaware to Aware
Aware to Interested
Interested to Considering
Considering to Intent
Intent to Action
Action to Loyalty
Agentic AI is now capable of detecting and predicting these shifts, even when the customer performs no explicit action.
If a subscriber moves from Aware to Considering because of repeated exposure to brand stories, that is performance.
If someone consistently skims EV content and begins visiting the website a week later, that is performance.
If a customer remains subscribed, opens regularly, and retains brand familiarity, that is performance.
The value is in the progression, not the click.
What This Means for Email Evaluation
Traditional KPIs break down quickly when applied to this new model.
A newsletter with minimal clicks but strong opens might drive a measurable lift in downstream site visits or vehicle research.
A product update with no CTA engagement might still trigger higher content affinity scores in the AI model.
A multi-step nurture sequence might move a customer from passive interest to active consideration long before they ever click anything.
When AI can interpret signal strength on its own, optimizing for clicks becomes counterproductive.
It encourages marketers to chase activity instead of intent.
The goal of CRM is not to generate actions inside the email.
The goal is to shape the customer’s trajectory.
The New Role of Email in an AI-Driven CRM Program
Email is no longer an endpoint. It is an input into an ecosystem.
Its purpose is to:
Provide consistent brand presence
Build mental availability
Reinforce key narratives
Increase affinity indicators that AI models can learn from
Support progression across the relationship curve
Sometimes that progression is obvious and measurable.
Often it is quiet and cumulative.
But it is still real, and AI can detect it.
The emails that create the most long-term value are often the ones that produce the fewest clicks.
The Takeaway
Clicks were useful when they were all we had.
Now they represent only a fragment of the picture.
In modern CRM:
AI detects engagement without explicit action.
Customer state changes matter more than in-email behavior.
Email success is defined by its influence on progression, not by its ability to drive clicks.
The most effective CRM programs optimize for trajectory, not interaction.
The question is no longer, “How do we increase clicks”
It is, “How do we design messages that move customers forward in ways AI can understand”
Not every email needs a click. In fact, the emails that shape the relationship most effectively may never need one at all.
The TRAP OF Hyper-Personalization (And How to do it right in 2025)
Written by Christine Harris
Personalization is one of the most overused words in marketing. In every strategy session, someone will say: “We should personalize more.” And they’re right… in theory. Experienced marketers know personalization is a driver of business success. But many don’t understand how to do it effectively, especially in CRM.
Here’s the reality:
Today: In most organizations, personalization is still heavily manual. A strategist decides which detail to highlight, and execution teams scramble to fetch the data, build queries, and create custom templates. The effort is high, and often misapplied.
Tomorrow: With unified profiles and agentic AI models (AI systems that can independently adapt and execute personalization logic across channels), the manual lift essentially disappears. You can personalize anything at scale, and in real time.
And that’s why strategy will always matter. Whether personalization is manual or automated, the key question isn’t what can we personalize? It’s what should we personalize to meaningfully move the customer forward?
The Pitfall of “Personalize Everything”
Consider healthcare.
A telemedicine provider has the ability to personalize before, during, and after each visit: pulling from past consultations, current symptoms, and even local seasonal health trends to recommend relevant follow-up care. Done well, this type of personalization guides the patient toward a meaningful next step, like booking another appointment or adopting a tailored wellness plan.
But here’s the caution: healthcare information is highly sensitive. Dropping in hyper-granular details (“We noticed you reported nausea at 8:37 p.m. last Thursday”) doesn’t advance the patient’s journey, it just feels invasive.
The same principle applies in CRM. Personalization that doesn’t move the customer forward in their funnel is wasted effort. Whether it’s a timestamp, an irrelevant browsing detail, or a room number from a past stay, novelty doesn’t equal impact.
How to Use Hyper-Personalization the Right Way (Today)
Until companies have advanced personalization engines in place, most decisions are still human-driven. That makes it even more important to evaluate whether a personalization detail is worth the lift.
Here’s the framework I use:
Relevance > novelty
If a detail doesn’t change how the customer thinks, feels, or acts, it’s clutter.Behavior + context
Not every action signals intent. Ten page views may be curiosity; a click on “Build & Price” is true interest.Profile + behavior = meaning
Combining who they are (e.g., prospect, loyalist, geography) with what they’ve just done creates actionable relevance.Experience vs. effort trade-off
If the manual lift outweighs the business value, it’s not worth it.
Start Simple: Small Data, Big Impact
You don’t need 20+ data points to make personalization effective. Even small touches make a difference:
Using a first name.
Pivoting the message slightly for new customers versus long-time fans.
Referencing a customer’s last interaction when it directly connects to their next step.
Past research by HubSpot has shown that personalized CTAs convert 202% better than generic ones. The point isn’t to chase every possible datapoint. It’s to use the right ones purposefully.
Tomorrow’s Opportunity
So what happens when AI takes over? With agentic models and unified data profiles, the manual work of execution disappears. A system can dynamically adapt messages to every nuance of a customer’s behavior.
But here’s the catch: even when personalization is technically limitless, the need for oversight and strategy doesn’t go away. If AI personalizes irrelevant details, you’ll simply end up with clutter at scale. Noise is still noise even if it’s automated.
Bridging the Gap
Most companies are somewhere in between. They want hyper-personalization, but their data, integrations, and tech stack aren’t fully ready.
The smart move in this stage is strategic restraint:
Personalize the details that matter most right now.
Use first-party data (e.g., lifecycle stage, product interest) to drive relevance.
Build the foundations — clean data, unified profiles, strong reporting — so that when AI enters the picture, personalization can scale meaningfully.
The Takeaway
Hyper-personalization is both a present challenge and a future opportunity.
Today, it’s about humans choosing wisely which details justify the effort.
Tomorrow, AI will remove a lot of the manual lift, but not the need for disciplined strategy and oversight.
So the guiding question remains timeless:
“Will this personalization make the message more relevant in a way that changes customer intent or drives measurable impact?”
If yes, invest in it. If not, step back. Sometimes the most sophisticated strategy is knowing when to keep it simple.
5 Tips for Effective Lifecycle Marketing in 2025
Written by Christine Harris
In 2025, customers expect personalized, connected experiences at every stage of their journey. At its core, lifecycle marketing builds on each step that came before it, guiding customers through the purchase cycle toward conversion and loyalty. A strong lifecycle strategy should feel seamless, timely, and personalized to the individual.
To achieve this, marketers must navigate evolving technology, stricter privacy rules, and the rise of AI. Lifecycle programs need to be both smarter and more disciplined. While execution looks different depending on purchase stage, there are a few fundamentals that apply universally.
There’s no one-size-fits-all approach to lifecycle marketing, but there are guiding principles that can make a real difference. I’ve shared a few of my own learnings here in the hope they’ll support you on your journey.
1. Start with clean, connected data
Strong lifecycle programs are only as good as the data behind them. The customer data you collect should tell the story of behaviors, preferences, and patterns — insights that ultimately guide your strategy. If that data is fragmented, outdated, or siloed, the customer experience will feel the same way.
The first priority is data hygiene and connection. Unified profiles, normalized fields, event tracking, and clear governance create the foundation for reliable journeys. Without this step, even the best creative or automation won’t deliver consistent results.
And be deliberate about defining what matters: viewed product, added to cart, completed “build & price,” or drop-off.
2. Focus on high-value personalization
The days of one-size-fits-all campaigns are over. With customer data, you can refine your audience with precision — not only to pinpoint lifecycle stage, but to tailor messages to the preferences of each customer group.
That said, it’s easy to overdo it. The goal isn’t to personalize everything; it’s to personalize the moments that matter. Look for signals of intent. A customer who clicks “Book a Test Drive” is telling you something far more valuable than one who simply browses a gallery page. By combining profile data with behavioral data, you can deliver personalization that is timely, relevant, and impactful.
3. Close the loop on reporting
Even strong lifecycle programs risk being undervalued if you can’t prove their impact. I’m often asked, “Which KPIs matter most?” Metrics like CTR, CTOR, and open rate all have their place, but they don’t tell the full story. A click is only valuable if you know what happened next. Did it drive a conversion? Was there immediate drop-off? How should you respond?
Reporting must connect lifecycle engagement to revenue, retention, and loyalty. Build dashboards that answer executive-level questions — demonstrating how your strategy supports sales and long-term growth. When you can tell that story, lifecycle marketing earns its place as a true business driver.
4. Test, learn, and evolve
Journeys are not “set it and forget it.” The best programs are living systems that adapt based on results.
All of the reporting work from tip three should fuel this step. On a macro level, lifecycle marketing is about continuously deepening the customer relationship. That means understanding what works, and what doesn’t.
Set aside a test-and-learn budget to experiment with content, cadence, and even AI-driven tools. Run A/B tests, act on the insights, and make iterative improvements. Small, continuous refinements compound into major gains over time.
5. Sequence advanced tactics wisely
AI, predictive scoring, and personalization engines are reshaping lifecycle marketing. These tools can unlock incredible opportunities — but only when the foundations are in place.
Think in phases: start with data, then establish reporting feedback loops, then scale journeys based on those insights, and finally layer on advanced AI tactics like predictive modeling or generative content. These tools can elevate personalization and meet customers more effectively at each stage of their purchase cycle. But if you leap ahead without the basics, you risk fragmented experiments instead of scalable results.
Bringing it all together
Effective lifecycle marketing in 2025 is about balance. New tools and tactics are exciting, but sustainable growth comes from pairing innovation with discipline.
If you’re leading lifecycle programs today, first ask yourself: Are we prioritizing the right foundations? or are we chasing features for the sake of it?