What Clay Is aware of That You Do not [+ Video]

Your advertising technique could also be caught in 2025.

Whilst you’re rigorously crafting segment-based campaigns, your rivals could be deploying AI brokers that know precisely when your prospects are prepared to purchase — generally even earlier than they do.

“Advertising based mostly on segments is turning into outdated,” says Alex Lindahl, go-to-market (GTM) engineer at Clay. On this unique dialog, he challenges standard advertising knowledge and introduces a revolutionary strategy that is producing 45% reply charges on chilly outreach. He additionally reveals how AI orchestration is remodeling buyer engagement from a scheduled broadcast into an clever dialog.

As somebody deeply concerned in advancing these advertising frontiers, Alex can be set to share his insights and hands-on methods throughout an upcoming workshop on the AI in Motion roadshow in Atlanta in June 2025.

This is not simply one other AI suppose piece — it is a sensible roadmap for creating what Clay calls “Go-to-Market Alpha,” the aggressive edge that comes from continuous experimentation and adaptation.

This interview is a part of G2’s Q&A sequence. For extra content material like this, subscribe to G2 Tea, a e-newsletter with SaaS-y information and leisure.

To observe the total interview, try the video beneath:

Contained in the business with Alex Lindahl

Many entrepreneurs are drowning in information however nonetheless fighting actionable insights. In your expertise working with Clay’s shoppers, how does an AI-first strategy basically change the best way entrepreneurs ought to take into consideration their buyer information? 

When interested by information utilization, it is vital to think about the various kinds of information which might be on the market. Now we have structured information, present in databases, CRMs, and product utilization metrics. Then there’s unstructured information scattered throughout the online within the type of weblog posts, job postings, monetary reviews, and numerous different sources. So it’s worthwhile to acknowledge that every firm has a singular best buyer profile (ICP) and distinct personas that require particular information units. So the hot button is determining the place you may supply that information, each internally and externally.

With the appearance of AI, we are able to now deploy brokers to assemble unstructured information from across the internet, analyze it successfully, extract significant insights, and develop actionable suggestions. The actual energy comes from bringing these two components collectively and mixing numerous information sources with AI capabilities to speed up understanding and resolution making.

Traditionally, information utilization has largely been handbook, usually restricted to fundamental segmentation for strategic selections. However at present, we are able to apply AI throughout totally different elements of the GTM funnel and enterprise processes to dramatically enhance how we collect data, analyze patterns, and take motion based mostly on complete information insights.

Provided that the hole between producing insights and truly implementing them is a continuing problem in advertising, what frameworks do you suggest for organizations to show AI-generated insights into concrete enterprise actions systematically? Are there particular instruments or processes that you have seen work significantly effectively?

Yeah completely. We consider Clay as a system of motion. So inside Clay, you may have totally different columns powered by AI or information sources, and AI can analyze and construction this information successfully. As an example, if you happen to’re concentrating on a VP of Advertising, AI can align the proper case research or weblog publish particularly for that particular person. The hot button is making use of AI in a structured approach throughout totally different workflow varieties.

AI brokers additionally come into play right here. Traditionally, ICP personas, and advertising and gross sales performs have existed individually and not using a cohesive timing ingredient connecting them. What I am now calling “best buyer alerts” defines particular indicators that reveal potential enterprise alternatives or challenges the place your product can present worth. This strategy strikes us away from conventional segments towards signal-based engagement.

And so I got here up with this framework referred to as “DRIVE” that buildings this strategy.

D – Detect: Establish the sign and decide the right way to detect it and the way continuously to seek for it.

R – Overview: Analyze what the sign implies. You possibly can leverage AI right here — for instance, if an organization is hiring DevOps engineers, AI may also help conclude that they are scaling their engineering staff. This assessment helps decide whether or not the sign warrants motion.

I – Establish: Uncover challenges the corporate might face in reaching its targets. AI excels right here when skilled by yourself information, weblog posts, case research, and their outcomes.

V – Worth: Align your answer’s worth to the precise scenario the corporate is experiencing.

E – Execute: Take acceptable motion based mostly on the sign.

The perfect implementation entails always-on alerts always monitoring outlined account lists. This lets you goal particular accounts exactly when related alerts emerge, indicating well timed alternatives for engagement.

Many entrepreneurs are experiencing “immediate fatigue” from frequently refining their AI directions. What frameworks or approaches do you suggest for creating extra sustainable, reusable immediate methods?

I personally discover immediate engineering extremely enjoyable and highly effective. Particularly as a non-technical particular person, as a result of now mainly everybody’s turning into an engineer. You should utilize platforms like Lovable to even immediate your method to creating an software. However within the GTM world, there are tons of prompts which you could create, tons of prompts which you could motion. 

So, how do you go from simply taking a immediate and refining it over time?

There are a number of approaches to create repeatability together with your prompts. One technique is creating a customized GPT by embedding your refined immediate right into a specialised mannequin, enriching it with contextual supplies and studying assets. This creates a reusable device that constantly generates advertising belongings or supplies aggressive evaluation with out requiring you to recreate the immediate every time.

One other implementation technique entails AI brokers. At Clay, we make the most of an AI agent that executes customized prompts for analysis functions. As an example, you may direct the agent to find out whether or not a particular firm has a purchasing cart on its web site. The agent will go to the positioning, analyze it, and return with a definitive reply or related clarification. 

The actual energy comes from making use of this functionality throughout a number of accounts — whether or not a whole bunch, hundreds, and even a whole bunch of hundreds — permitting you to scale your immediate’s performance effectively.

The third strategy integrates these prompts immediately into operational workflows. Think about a easy instance: When an inbound lead arrives, you may apply a immediate to categorize the particular person’s title based mostly on seniority or job perform. A subsequent layer may then suggest probably the most acceptable case research to ship them. This creates a sequence the place AI interventions result in progressively higher information or motion outputs.

The method requires some preliminary funding in creating, refining, and testing your prompts. Nonetheless, as soon as optimized, you may deploy them by way of customized GPTs, AI brokers, or workflow integrations the place they will perform repeatedly inside their designated context. This eliminates the tedious cycle of repeatedly copying and pasting prompts into ChatGPT, which shortly turns into exhausting.

Your “Construct the Prompts & Motion the Insights” session within the “AI in Motion Roadshow” guarantees to assist entrepreneurs rethink their present strategy. What transformation do you usually see in entrepreneurs’ considering from the start to the top of this workshop?

I believe what lots of people are realizing is that advertising based mostly on segments is turning into outdated. The brand new strategy is utilizing alerts to align the proper advertising or gross sales play to your ICP or persona. So, we’re seeing a shift in considering — alerts are actually taking priority over segments.

Reasonably than creating standard campaigns the place you may ship one message this week and a unique one two weeks later with maybe a couple of variations, we’re now focusing extra on listening to the market, accounts, or people. This enables us to use actions or ship advertising messages exactly once they’re almost certainly to resonate with the recipient. That is one of many key insights folks usually take away from our workshop.

One other realization is that immediate engineering, whereas sounding advanced or intimidating, is definitely fairly accessible. You solely want to grasp some fundamentals to enhance your prompts over time. You possibly can start with a fundamental immediate and easily ask AI the right way to improve it or what components could be lacking. This back-and-forth dialogue with the AI helps refine your immediate based mostly on its suggestions, and you’ll incorporate these ideas and even have the AI rewrite the immediate totally for you.

You simply must ask the proper questions and take into consideration what the output is that you really want, so you may articulate that clearly to AI. And maybe probably the most vital takeaway is discovering how simply you may create a customized GPT, apply your refined immediate, and use it constantly to make that immediate repeatable.

This additionally accompanies the broader notion that everybody is actually turning into an engineer at present. You possibly can create purposes, construct merchandise, present AI directions, and develop AI brokers — capabilities that weren’t beforehand accessible to non-technical folks. 

It is thrilling that people with out technical backgrounds can now carry out extra technical capabilities by leveraging AI, making the whole course of not simply productive however fulfilling as effectively.

Alex Lindahl
GTM Engineer at Clay

Your workshop covers “AI orchestrations.” Might you clarify how this idea particularly applies to advertising workflows and supply an instance of how this has improved marketing campaign efficiency?

So, plenty of the orchestrating of AI brokers might be achieved by way of a number of approaches. In Clay, it is fairly simple — you arrange a column that applies your AI agent directions repeatedly throughout an account checklist. Alternatively, you may deploy an ambient agent that displays for particular alerts or implement a chat agent that engages with prospects to qualify leads.

Certainly one of my favourite examples entails configuring a customized sign to trace sure key phrases on LinkedIn. When these key phrases seem in posts, each the content material and poster are routinely imported into Clay. The system then makes use of AI to investigate the particular person’s title, persona, and message content material. Primarily based on this evaluation — significantly specializing in persona and title — it determines whether or not to proceed with message creation or take no motion.

The following section entails AI analyzing the publish and finishing a easy template like, “I simply noticed your publish on [blank].” with AI filling in that clean house. This creates a customized snippet which you could incorporate into your outreach. From there, an automation can set off both a connection request, a LinkedIn message, or an e mail to that particular person.

For this explicit marketing campaign, I really noticed a exceptional 45% reply price, which is totally insane by business requirements. This success stems from incorporating timing components and just-in-time messaging through the use of the detected sign to find out the optimum second for outreach. In fact, we’re leveraging AI to filter prospects and guarantee we’re connecting with the suitable seniority ranges.

As you may see, this strategy considerably boosts the general marketing campaign effectiveness as a result of it is extra personalised, well timed, and basically extra related to what that particular person is considering or experiencing in that particular second.

For individuals who acknowledge the potential of an AI-first strategy however work in conventional organizations, what sensible recommendation would you give to somebody who needs to champion these strategies inside their staff however faces institutional inertia or skepticism from management?

I’d say first develop some degree of consolation and information round AI itself. Take a activity you imagine might be automated and experiment with instruments like ChatGPT or a customized GPT to streamline it. This preliminary step focuses on enhancing your individual productiveness whereas gaining priceless prompting expertise and understanding what’s potential and what is not.

Via this course of, you will naturally change into extra environment friendly and develop shortcuts for particular duties, reminiscent of changing name transcripts into follow-up emails. This hands-on expertise builds your basis in immediate engineering.

Subsequent, establish a mission inside your staff the place you may create a immediate that is going to assist speed up that mission and transfer it ahead. And thru that, you are going to begin to expose your teammates to prompting and AI, they usually’ll in all probability be fairly stunned. As soon as you have established this foothold, decide the right way to combine AI into your staff’s common operations — their each day or weekly workflows. This gradual growth demonstrates worth at more and more broader ranges: first private productiveness, then mission acceleration, and eventually, team-wide implementation.

These progressive successes present the proof wanted to construct a compelling enterprise case. Now you can make knowledgeable suggestions based mostly on demonstrated outcomes: “We have efficiently utilized this at particular person, mission, and staff ranges. Let’s scale this strategy to different departments to attain comparable return-on-investment (ROI), productiveness features, and efficiencies.”

If readers might take away only one perception from this text, what would you need them to recollect?

I’d say a very powerful takeaways are actually two issues. First, have enjoyable with AI as a result of AI exploration might be genuinely fulfilling. Second, experiment with your individual use instances and inside your organization’s GTM technique.

At Clay, we have developed this idea of “Go-to-Market Alpha,” which is actually discovering methods to do issues in a different way to realize aggressive benefits. You possibly can’t uncover these benefits with out experimenting. Some experiments will not present ROI, however by way of this course of, you will uncover approaches that work remarkably effectively.

The previous playbooks aren’t working as a result of we have stopped experimenting. We have to frequently experiment to seek out that new playbook, that new wedge, that new ‘Go-to-Market Alpha’.

Alex Lindahl
GTM Engineer at Clay

So maintain experimenting each personally and together with your staff and firm. You will begin to uncover totally different pockets of “alpha” — aggressive benefits that profit each you and your group. 

If you happen to loved this insightful dialog, subscribe to G2 Tea for the newest tech and advertising thought management.

Observe Alex Lindahl on LinkedIn to study extra about sign based mostly advertising.

---

Edited by Supanna Das