Most guides on finding B2B leads with AI describe the same thing: pick a data provider, filter by industry and company size, export a list, and hand it to a sequencer. That is database querying with an AI label on it. It is not what an AI agent does.
An AI agent starts from your offer. You describe what you sell and who needs it. The agent reads that description, searches for profiles that match it, and tells you for each one whether it qualifies and why. The output is not a list you still have to clean. It is a scored pipeline, built from your brief.
The difference matters in practice. A query that returns “VP Sales, 50-200 employees, SaaS” gives you everyone who fits the filter. A brief that describes “I help Series A SaaS companies reduce churn by redesigning their onboarding flow” gives the agent grounds to distinguish a VP Sales who owns onboarding from one who does not, and to explain the difference. That reasoning is what turns a database search into a lead list worth acting on.
Step 1: Write a Brief the Agent Can Reason From
The quality of your lead list is determined before the agent runs a single search. It is determined by the quality of your brief.
A brief has two parts: the offer and the persona.
The offer is not your company description. It is the specific problem you solve, for whom, and what the outcome looks like. “B2B consulting” is a category, not a brief. “I help HR directors at companies going through a merger restructure their management layer before headcount decisions are made” is something an agent can match against a profile. The more specific the problem, the more useful the scoring will be.
The persona defines your target beyond job title. Title is the starting point, but it is not the filter. Two HR directors at structurally identical companies can be at completely different points in their decision cycle. The persona captures the signals that distinguish a relevant profile from a technically matching but useless one: the type of company, the growth stage, the challenges that indicate the problem you solve is live.
If your ICP is not fully defined yet, work through it before running discovery. My guide on how to define your ICP with AI covers the process in detail. Running discovery with a vague brief is not faster than running it with a precise one. It just produces a list you will have to filter by hand, which removes the point of using an agent.
Step 2: Let the Agent Source and Score Profiles
With a brief in place, the agent starts searching. This is where the behavior diverges sharply from a database export.
A database filter returns every record that matches a declared field. An AI agent reads each profile the way a human reviewer would: it interprets job titles, responsibilities, stated missions, and company context against your offer and persona. It then assigns a score from 1 to 5 and writes a justification.
The score means something specific. A 5-star profile matches both your offer and your persona criteria closely. A 3-star profile meets the threshold but with notable gaps (perhaps the role fits but the company is outside your target sector). A 2-star profile does not qualify, but the agent surfaces it anyway so you can override manually if you see something it missed.
The justification is as important as the score. Reading “3/5: relevant role, but company too early-stage for the offer” tells you something a filter never could: this profile was considered, not skipped.
In LEO, prospect discovery works exactly this way. I search LinkedIn profiles one by one against your offer and persona. For any profile scoring 3/5 or above, I retain the prospect and present it for your review or, in Semi-Auto mode, add it automatically to your pipeline. You can reject a profile and it will not be proposed again. If I cannot find enough qualified profiles for your persona, I tell you and suggest adjusting the criteria.
Semi-Auto handles the sourcing loop without requiring you to be present: I analyze profiles continuously, apply your scoring threshold, and build your prospect list in the background. You then manage contact actions once the list is ready. For the full picture of how the prospect cycle runs from discovery to outreach, my guide How to Prospect with AI: A Step-by-Step Process covers each stage.
Step 3: Read the Score, Not Just the Number
A scored list is only useful if you know how to interpret it.
The most common mistake is treating the score as a pass/fail gate. A 3/5 prospect that happens to be a warm referral from a mutual connection is more valuable than a 5/5 cold contact you have no route into. The score reflects fit against your brief. It does not factor in context you hold but the agent does not.
Two things to check when reviewing a scored list. First, look at the score distribution. If 80% of your profiles are landing at 3/5, either your persona criteria are too broad or your offer description is too generic for the agent to discriminate effectively. A tighter brief produces a distribution where 4s and 5s are common. Second, read the justifications on the low-scorers before discarding them. Occasionally, a 3-star profile has a gap that does not matter for your specific outreach angle. The justification tells you whether the gap is material.
For a deeper look at how the scoring mechanism works internally, calibration options, and how to adjust thresholds as your pipeline grows, the dedicated guide on AI lead scoring for B2B covers that level of detail.
Step 4: Enrich Before You Contact
A qualified lead is not a contactable lead. Before the agent generates a message, it needs to know enough about the prospect to make that message relevant.
Enrichment is not about collecting data for its own sake. It is about giving the agent what it needs to write a message that does not read like a template. A message generated without enrichment will be accurate about job title and company name and wrong about everything that matters: the challenges the person is dealing with right now, the relevant context from their recent activity, the signal that makes the outreach timely.
What I enrich: identity, current role, responsibilities, areas of expertise, stated interests, identified challenges, and company context including size, estimated revenue, sector challenges, and recent news. This enrichment covers professional profile and company data. Email and phone are found in a separate step when you are ready to reach out, triggered when the contact stage begins.
The rule is straightforward: enrich first, contact second. A list of 50 well-scored but un-enriched profiles will produce worse outreach results than a list of 20 enriched ones. The enrichment step is where discovery becomes actionable.
For the full method on what to enrich and how to use enrichment data in message generation, my guide on B2B lead enrichment goes into that detail. The point here is that enrichment is a prerequisite, not an optional extra.
Step 5: Detect When the List Is Off-Target
A well-configured agent is not immune to calibration drift. Over the first 20 to 30 prospects discovered, patterns emerge that tell you whether the brief is working as intended.
Three signals indicate the list is too broad or mis-calibrated.
First: score distribution skewed low. If the majority of retained prospects are scoring 3/5, the agent is qualifying on minimum criteria. The brief is matching profiles on surface indicators (title, company size) rather than on the specific problem your offer solves.
Second: consistent sector or role drift. If you defined a persona targeting VP Operations at logistics companies and your list fills up with Operations Managers at SaaS companies, the persona criteria are either ambiguous or in conflict with each other. An agent will optimize toward whatever fits most frequently, which is not always what you intended.
Third: justifications that repeat the same gap. If 60% of your 3-star profiles show the same missing criterion (wrong company stage, wrong industry, mismatched seniority), that criterion is a systematic filter, not an edge case. Either tighten the persona to exclude it up front, or lower its weight if it genuinely does not affect fit.
None of these require starting over. Adjusting the offer description to be more specific, narrowing the sector list in the persona, or raising the minimum score threshold from 3 to 4 will shift the distribution within the next discovery cycle. The goal is a list where most profiles are genuinely worth the effort of enrichment and outreach, not a list that needs another round of manual filtering before it becomes usable.
If you are running discovery across several personas simultaneously, read calibration signals at the persona level, not the account level. A drift in sector fit or a low-score concentration that looks minor in aggregate can point to one specific persona that is too broadly defined. Correcting it individually keeps the other personas unaffected.
Step 6: Measure Lead Quality, Not List Size
List size is a lagging indicator. It tells you the agent ran, not whether it found the right people.
Two metrics tell you whether your lead generation is working: the proportion of 4- and 5-star profiles in your active pipeline, and the reply rate on first-contact messages before any follow-up is sent.
If high-scoring profiles make up a small fraction of your pipeline, your brief is producing a lot of technically qualifying profiles that do not actually fit. If your first-message reply rate stays below 5%, the combination of targeting and messaging is off, and the most likely culprit is the targeting, not the message.
Both metrics improve together when the brief is precise. A narrow, well-defined offer description produces higher-scoring leads, which produce more relevant messages, which produce better reply rates. The reverse is also true: a vague brief produces low-scoring leads, generic messages, and a pipeline that looks active but generates no replies.
For a full breakdown of which metrics to track across the prospecting cycle and how to interpret them in context, my guide on B2B AI lead generation covers the measurement layer in detail.
If you want to run this process now, run your first discovery cycle. The first session is the onboarding: I collect your business context, build an offer and persona with you, and run the first discovery cycle so you can see a scored list before the trial ends.
Common Mistakes to Avoid
Starting discovery before the offer is specific enough
The most frequent calibration failure is not a configuration problem. It is a brief problem. If the offer description describes a category rather than a problem, the agent cannot discriminate. Fix the brief first.
Reading only the score, skipping the justification
The justification is where the diagnostic information lives. A consistent gap in low-scoring profiles is a signal about your brief, not a noise to dismiss.
Skipping enrichment to contact faster
A 50-profile list enriched and contacted one by one will outperform a 200-profile list contacted on title and company name alone. Speed in outreach is not volume; it is the right message at the right time to the right person.
Treating list size as success
A pipeline of 300 leads that generates no replies is not an asset. It is evidence that the discovery is off-calibrated. The right question after building a list is not “how many?” but “what is my 4-5 star rate, and what did my first messages produce?”
Activating automated outreach before validating the first manual cycle
Running discovery and reviewing the first ten scored profiles manually tells you whether the brief is calibrated. Activating automation before that check means scaling a calibration error rather than a working process.
