TL;DR
- The Shift: Real-time intent beats a purchased list every single time.
- The Signal: Comments, follows, and community posts tell you who is in market right now.
- The Speed: Outreach within hours of a signal converts because the pain is still fresh.
- The Trap: Data without routing is just a faster way to annoy people.
Most demand generation still runs on a list. You buy a database, load it into a sequencer, and hope that a fraction of the names are in market for what you sell. It is not that the list is wrong. It is that the list is deaf. It knows a person’s title and company, and it knows nothing about what that person cares about this week.
There is a better input, and it is public. Your buyers are already telling you what they care about, in the places they hang out, hours before anyone makes a buying decision. The work is not finding them. The work is building the system that notices.
Intent Is a Behavior, Not a Filter
When someone comments on a competitor’s post, they are naming a problem in front of an audience. When they follow a niche industry account, they are telling you which category they are shopping. When they post about a specific frustration, they are writing your outreach for you. None of that is a filter you set in a database. It is a behavior you observe.
The distinction matters because filters describe a static target and behavior describes a moving one. A title in a CRM is true until the person changes jobs. A public signal is true for the length of a news cycle, and that is exactly the point. It is perishable, which is what makes it valuable.
This is where most intent data vendors lose the plot. They sell you a score, a number between one and a hundred that is supposed to mean urgency. But a score is a summary of behavior, and summaries are always late. The raw behavior is the asset. The moment you compress it into a score, you have thrown away the thing that made it useful, which is the specific words the buyer used.
Speed Is the Whole Advantage
An intent signal has a half-life. Someone who posts about a broken process on a Tuesday is most reachable on Tuesday, and progressively less reachable every day after. By the following week, either they solved it, or they stopped thinking about it, or someone faster got there first.
So the system has to be fast, and speed in this context is not about working harder. It is about removing the human from the loop between signal and outreach, while keeping the human in the loop on anything that actually goes out. The machine notices. You decide what to say. That division of labor is the entire trick.
I built the routing so that a fresh signal creates a task, not a message. The task carries the context: what the person said, where they said it, which of my offers fits. Then a person, or an agent with approval, turns that context into a note that reads like it was written by someone who was paying attention, because it was.
There is a reason this beats a polished sequence. A sequence is optimized for volume. A signal response is optimized for relevance, and relevance is the only thing that survives a crowded inbox. You would rather send forty messages that land than four hundred that get filtered into oblivion.
Data Without Routing Is Just Noise
Here is the failure mode I see most often. A team buys an intent tool, gets a firehose of signals, and does nothing with them because there is no system to act on the data. The tool becomes an expensive dashboard nobody opens. The problem was never the data. It was the absence of a workflow downstream of it.
Building the workflow is the actual work, and it is mostly unglamorous. You have to decide which signals matter for which offer, who owns the response, what the response looks like, and how you know it worked. Skip any of those and you have a faster way to notice that you are not doing anything about what you noticed.
Your best leads are not in a database. They are commenting on your competitor’s post right now, describing the exact problem you solve. The question is not whether the signal exists. It is whether anything in your business is built to hear it.
The routing question is also the design question. If every signal pages you, you will build an alert-fatigue machine and ignore it within a week. If no signal pages you, the data is decorative. The useful middle is a small set of signals that map cleanly to offers you actually sell, bound to a person who owns the follow-through.
Where Buyers Actually Signal
Signals live wherever your buyers hang out, and that is rarely one place. It is professional networks, niche communities, industry accounts, and the comment sections where practitioners argue about how the work is done. Each surface has a different texture, and each one rewards a different kind of response.
| Surface | What the signal tells you | Right response |
|---|---|---|
| Competitor comments | They are evaluating the category now | Lead with the gap the competitor leaves |
| Industry follows | They are tracking a theme | Share a point of view on that theme |
| Pain-point posts | They named the problem in their words | Echo their language back to them |
| Community threads | They are asking for a solution publicly | Answer helpfully, then offer next step |
The table is not a playbook to copy. It is a demonstration that the response has to match the signal, because the signal is a clue about where the person is in their thinking. A follow is early. A post is mid. A comment on a competitor’s launch is late and warm. Treating them all the same is how you waste the data.
What I Actually Think
Most teams are one workflow away from a much better pipeline, and they keep buying inputs instead of building the workflow. Buying another data source is easy. Building the routing, the ownership, and the feedback loop is work, which is exactly why it is a durable advantage once it exists.
I would rather have a crude signal feed wired into a working routing system than a perfect data set that lands in an unread dashboard. The value is in the last mile, not the first. This is the same discipline behind the signal-first framework I use for content, and it is why the agents running my marketing ops are wired to notice and route rather than to spray. The system exists to close the gap between what your buyers say in public and what your business does about it.
Start smaller than you want to. Pick one signal that matters for one offer, and build the route from noticed to sent. If it works, the signal has earned the next one. If it fails, you have learned something cheap. Either way, you stop guessing about who is in market, because now you can hear them.
The other thing I would say, and I have learned this the hard way, is that the feed gets quieter before it gets louder. The first week of watching a public surface, you will see maybe three signals worth acting on, and it will feel like the data is thin. Resist the urge to widen the net. What is actually happening is that you are calibrating your sense of what counts, and that calibration is the skill the system depends on. A wide, noisy feed on day one teaches you nothing and buries the good signals in the bad ones.
Once the calibration is there, the volume problem solves itself, because you can now point the same machinery at a second surface without drowning. The mistake is to start wide and hope the tooling sorts it out. The tooling is only as good as the definition of a signal you bring to it, and that definition only comes from doing the narrow version first.
Pick one surface where your buyers complain in public. Watch it for a week and write down every signal that maps to an offer you sell. That short list is your first intent feed. The tooling comes after, not before.
Turn signals into conversations →
