What Is Sales Prospecting? The 2026 B2B Guide to Finding Buyers

What Sales Prospecting Actually Is

Sales prospecting is the process of identifying potential buyers, researching whether they fit your ideal customer profile, and initiating direct contact to create a qualified sales conversation. The word borrows from mining: a prospector surveys ground that might contain something valuable, samples it cheaply, and only commits real effort where the evidence is strong.

That metaphor holds up better than most sales jargon. Prospecting is not selling. It is the disciplined act of deciding where not to dig. The rep who books eight meetings a month and the rep who books two are rarely separated by charisma. They are separated by the quality of the list, the accuracy of the contact data on it, and the order in which they work it.

Three things distinguish prospecting from the rest of the sales process:

• It is outbound-initiated by definition. The prospect has not raised a hand. If they filled in a form, you are working an inbound lead, not prospecting.
• It happens before qualification, not after. You prospect to earn the right to qualify. BANT, MEDDIC and every other framework apply to the conversation prospecting produces.
• It is a data operation as much as a selling one. Every touch depends on a field in a record being present and correct: an email, a direct dial, a company, a job title.

Client prospecting: new logos vs. your existing base

"Client prospecting" is used two different ways, and the difference matters more than the label. Some teams mean prospecting for new clients, which is simply outbound by another name. Others mean prospecting inside accounts they already serve: a second department, a new subsidiary, a replacement buyer after the champion left.

The second is consistently cheaper and consistently under-worked. You already hold the domain, the account hierarchy and the relationship history. What you usually lack is the contact detail for the people you have never spoken to, because your CRM only holds the handful who happened to sign the last contract.

Practically, that makes existing-client prospecting an enrichment problem more than a targeting one. Pull the org chart for accounts you already own, identify the roles you have no relationship with, and resolve their contact details. The fit question is already answered, which removes the most expensive source of waste in new-logo prospecting.

Prospect vs. Lead vs. Opportunity: The Definitions People Get Wrong

These four words get used interchangeably in most sales teams, which is precisely why forecasting conversations go badly. The distinction is not academic. It determines which stage a record sits in, who owns it, and what happens next.

Swipe the table sideways to see every column →

The line most teams blur is prospect. A name on a list is not a prospect. A name with no working phone number and a guessed email address is not a prospect either. It is a hypothesis. It only becomes a prospect when you can put a message in front of that specific human. This is not pedantry: it changes your list size, your ratios and your headcount plan, which we work through in the meeting maths section.

How does someone become a prospect?

Three qualification gates, applied in order. Each one is cheap relative to the next, which is why the order matters:

1. Fit. Does the company match your ICP on industry, size, geography, tech stack or whatever your real buying signal is? Answerable from firmographic data alone, at effectively zero cost per record.
2. Role. Is this person plausibly a buyer, a user, a budget holder or a blocker? Answerable from job title and department, again cheaply.
3. Reachability. Do you hold a verified business email or mobile number for them? This is the only gate that costs real money, and the only one most teams never measure.

Sales Prospecting vs. Lead Generation

Short version: lead generation is a system that makes buyers come to you; prospecting is a practice of going to buyers directly. They produce records that look similar in a CRM and behave completely differently in a pipeline.

Swipe the table sideways to see every column →

The two are complements, not substitutes. Lead generation gives you compounding, low-marginal-cost volume that you do not control precisely. Prospecting gives you precise control over exactly which 200 accounts you attack this quarter, at a cost that rises with every record. Teams that pick only one either cannot grow on demand (pure inbound) or cannot grow efficiently (pure outbound).

If you want the other side of this in depth, we cover it in What Is Lead Generation? A B2B Practitioner's Guide.

Is prospecting the first step in the sales process?

Yes. In the classical seven-step model (prospect, prepare, approach, present, handle objections, close, follow up) prospecting is step one, and in every modern variant it remains the first activity a seller performs against a specific named person.

The nuance worth holding: prospecting is the first step of the sales process but not the first step of the revenue process. Segmentation, ICP definition and list construction all precede it. Skip those and prospecting becomes an expensive random walk.

The Reachability Gap: Where Prospecting Actually Breaks

Ask a sales leader why prospecting underperformed last quarter and you will hear about messaging, cadence discipline, or a rep who "isn't hungry enough." Ask a RevOps analyst to pull the numbers and you will usually find something duller: a meaningful share of the list was never contactable in the first place.

Here is the sequence that quietly destroys outbound programmes:

1. Marketing and sales agree on an ICP. Say 2,000 accounts.
2. A list is built. Four relevant contacts per account, which is a conservative read of a typical B2B buying committee, gives 8,000 names.
3. The names are enriched. Emails come back for most, mobiles for far fewer.
4. Sequences launch against all 8,000, because the CRM does not distinguish between "no email" and "email we have not tried yet."
5. Activity metrics look healthy. Meeting counts do not move.

The failure is invisible in every dashboard a sales team normally looks at, because activity dashboards count attempts, not attempts that could physically have arrived. A call to a switchboard that never reaches the buyer is logged identically to a call that gets picked up. An email to a dead mailbox is logged as sent.

Model your own reachability gap

The figure below is a calculator, not a benchmark. It contains no claims about anyone's coverage. You set the inputs, and it does the arithmetic that most CRM reports quietly skip. If you do not know your real coverage numbers, that is itself the finding: run a sample of 500 records and measure it before you plan a quarter around it.

Two things fall out of this model almost regardless of the inputs you choose.

First, mobile is usually the swing variable. Not because email coverage is uniform across providers, it is not, but because email and mobile fail differently. An email gap is usually visible immediately, in the form of a bounce. A mobile gap is silent: the record simply has no number, or has a switchboard number that looks like data and behaves like nothing. Move the mobile slider from 20% to 60% and watch what happens to the reachable count. That delta is your entire multi-channel capability, and it is the number you should ask any provider to quote for your specific segment rather than globally.

Second, the unreachable segment is not free. Those records still consume list-building time, still occupy CRM licences, still get counted in "accounts covered" reporting, and still make your reply-rate denominator look worse than your messaging deserves. Removing them from the denominator is often the fastest way to find out whether you have a data problem or a messaging problem.

If you want the underlying mechanics of how coverage is produced and why providers disagree so wildly, see What Is Waterfall Enrichment? and What Match Rate Should You Expect From a Contact Enrichment API?.

The Sales Prospecting Process: Six Stages

There is no shortage of prospecting frameworks. Most collapse into the same six stages, and the only one that reliably differs between good teams and bad ones is stage three. Click any stage below to see what it involves, what it costs, and what happens when it is skipped.

Why stage 3 gets skipped

Because it is the only stage with a visible line item. Defining an ICP costs a workshop. Sourcing names is bundled into a platform subscription you already pay for. Sequencing is bundled too. Verification is the one step where somebody has to approve incremental spend, so it gets deferred, and the cost reappears later as low connect rates, high bounce rates and domain reputation damage, none of which are traceable back to the decision that caused them.

Prospecting Methods and Techniques Compared: What Each One Requires

Most comparisons of prospecting techniques argue about which channel "works best." That question has no general answer; it depends on your ACV, your buyer, and your market. A more useful question, and one that does have a stable answer: what data does each method require in order to function at all?

This matters because it converts a strategy debate into an inventory check. If 60% of your list has no mobile number, "we should do more cold calling" is not a strategy. It is a purchase order you have not raised yet.

Outbound, inbound and social prospecting

The three labels describe where the first touch originates, not how sophisticated the approach is.

LinkedIn prospecting: what it can and cannot do

LinkedIn is where most B2B prospecting research starts, and where a lot of it wrongly ends. The platform is excellent at three things: confirming that a person currently holds the role you think they hold, showing you who else sits around them in the buying committee, and surfacing trigger events such as a promotion or a company move.

It is deliberately poor at the fourth thing you need. LinkedIn does not give you a phone number, and in most cases it does not give you a work email either. Connection requests and InMail are rate-limited by design, so the channel has a hard ceiling on volume no matter what you spend.

The practical pattern that works:

1. Build the target list inside LinkedIn or Sales Navigator using role, seniority, headcount and geography filters. This is genuinely the best targeting interface in B2B.
2. Export the identifiers. A public profile URL is a stable, unambiguous key for a person, which makes it a far better enrichment input than a name-plus-company guess.
3. Resolve each profile to a verified business email and mobile through an enrichment step. This is what converts a research list into a workable one.
4. Run the outreach off-platform, using LinkedIn engagement as a warming layer rather than the primary channel.

Telephone prospecting

Cold calling did not die; it got harder to instrument. Two structural changes explain most of the difficulty. Buyers stopped sitting at desk phones, which killed the switchboard route to a named person. And mobile carriers began flagging high-volume unknown numbers, which suppresses pickup rates before a human ever decides whether to answer.

What follows from that is unglamorous but decisive. Telephone prospecting is now almost entirely a function of whether you hold a genuine personal mobile or direct dial for the individual. Rep technique matters at the margin. Number quality sets the ceiling.

Three checks before you scale a calling motion:

• Confirm what you are actually buying. "Phone" in a data schema often means the company reception desk. Ask the provider to split coverage into mobile, direct dial and switchboard, and price only the first two into your plan.
• Screen against national do-not-call registers before dialling. Obligations differ country by country and the liability sits with you, not the data vendor.
• Measure connect rate, not dials. Dials measure effort. Connect rate measures whether your numbers are real.

Email prospecting

Cold email remains the highest-throughput prospecting channel and the one most sensitive to data quality, because the damage from a bad list is not confined to the bad records. Hard bounces degrade your sending reputation, which suppresses inbox placement for the valid addresses in the same campaign and for weeks afterwards.

That makes verification a deliverability control, not just a coverage improvement. Three things are worth building into the process regardless of which sending tool you use:

• Verify at the point of send, not at the point of purchase. A record verified six months ago is an assumption, not a fact.
• Treat catch-all domains as a separate bucket. They accept everything at the server and tell you nothing, so they belong in their own low-volume segment rather than mixed into the main campaign.
• Exclude role-based addresses such as info@ and sales@ from person-level sequences. They inflate your list size and depress every ratio you use to judge it.

The channel-mix question, answered properly

You do not choose a channel mix. Your data chooses it for you, and you either accept that or you go and buy your way out of it. A team with 85% email coverage and 15% mobile coverage has an email-led motion whether or not that was the plan. A team that wants a call-led motion needs to fix mobile coverage first, and should treat "increase mobile coverage from X to Y" as a project with an owner and a cost, not an aspiration in a QBR deck.

How to Find Prospects and Build a Prospecting List

A prospecting list is a data product. It has inputs, a schema, a build process and a shelf life. Treat it that way and it stops being the thing your SDRs quietly spend two days a week rebuilding.

The minimum viable schema

Fields fall into three tiers. Tier 1 fields determine whether the record is workable at all; tier 2 determines whether it is worth working; tier 3 determines what you say.

Swipe the table sideways to see every column →

A build process that survives contact with reality

1. Define the account universe first, contacts second. Filter to companies before you spend a cent on person-level data. Account-level filtering is cheap; contact-level enrichment is not.
2. Deduplicate against your CRM before enriching. Enriching a record you already own, or one that is in an open opportunity, is pure waste. Worse, it produces the double-touch that annoys buyers.
3. Enrich in a sample first. Run 300–500 records before you run 30,000. Measure your actual match rate and, separately, your actual bounce rate. They are different numbers and vendors rarely publish the second.
4. Split the list by reachability, not alphabetically. Records with a verified mobile go to the call-led sequence. Email-only records go to the email-led sequence. Records with neither go back to enrichment or out of the campaign.
5. Set a refresh cadence. B2B contact data decays continuously as people change roles. A list that was accurate in January is not accurate in July, and the decay is fastest in exactly the high-growth accounts you most want to sell to.
6. Write back to the system of record. If enrichment output lives in a spreadsheet, you will buy the same record again next quarter. See CRM data enrichment for Salesforce, HubSpot and Dynamics 365.

Where to find prospects: the four sources

Four sources, in rough order of how much of your list they can realistically cover:

• Contact databases such as ZoomInfo, Apollo, Cognism and Lusha. Large static indexes you search and export. Fast, broad, and the coverage you get is whatever that single vendor happens to hold.
• Waterfall enrichment. A query is passed across many providers in sequence until one returns a verified match. Higher coverage than any single source, priced per result rather than per seat. Explained in full in our waterfall enrichment guide.
• Your own first-party data. Closed-lost accounts, churned customers, product sign-ups, event lists. Usually the highest-converting source available to a team, and usually the least systematically worked.
• Public and manual research. Company websites, filings, professional networks. Accurate, unscalable, appropriate only for a genuinely small target list.

For a like-for-like view of what the main platforms actually charge and return, see B2B Contact Data Pricing in 2026 and Top B2B Contact Enrichment API Providers.

AI and Agentic Prospecting: What Actually Changed

"AI sales prospecting" covers three genuinely different things that get sold as one. Separating them makes it much easier to work out what is worth buying.

Swipe the table sideways to see every column →

The third category is the one changing how prospecting stacks get assembled. Until recently, connecting a data source to a workflow meant a developer writing integration code against a REST API. The Model Context Protocol, an open standard for how AI applications talk to external tools, changed that. An agent running inside a client such as Claude, Cursor or a custom backend can now call an enrichment tool directly, in natural language, without bespoke integration work.

For prospecting specifically, that means the sequence "find the accounts, find the people, resolve their contact details, write them to the CRM, start the sequence" can run as one agent workflow rather than five tools and four exports. We compare what the major providers expose in B2B enrichment for AI agents: MCP server comparison.

Prioritising and Qualifying: Working the List in the Right Order

Once you have a reachable list, the next question is order. Most teams work lists alphabetically, by account name, or by whatever the export happened to sort by. This is a materially expensive habit: the first two weeks of a campaign are when rep energy, message freshness and manager attention are highest, and they get spent on an arbitrary sample.

The workable model uses two axes and nothing else.

• Fit. How closely the account matches your ICP. Static, computable from firmographics, and stable for months.
• Intent. Evidence that something is happening right now: a relevant hire, a funding round, a tech change, a website visit, a competitor renewal date.

What to do with each quadrant

Swipe the table sideways to see every column →

The quadrant nobody manages is the bottom-left. Leaving low-fit, low-intent records in an active campaign does more damage than ignoring them: it depresses every conversion metric you use to judge the campaign, which leads to the wrong diagnosis and the wrong fix.

Qualification: what to establish on the first call

Prospecting qualification is deliberately lighter than deal qualification. You are answering one question: is there a reason to have a second conversation? Four checks are enough:

1. Problem. Do they have the problem you solve, in a form they recognise?
2. Priority. Is it on a list of things they intend to do something about this year?
3. Person. Are they the buyer, an influencer, or neither?
4. Process. If they wanted to move, do they know how a purchase would happen?

Anything beyond this, including budget confirmation, technical validation and security review, belongs to the AE. Pushing full deal qualification into the prospecting call is the most common reason first meetings do not convert to second ones.

The Meeting Maths: How Many Prospects Do You Actually Need?

This is the calculation that turns prospecting from an activity into a plan. It is also the calculation that explains the most common complaint in outbound sales: "my BDRs are hitting activity targets and not booking meetings."

Work backwards from the meetings you need. Every stage multiplies, so a small change in the first term, the share of your list that is reachable at all, moves the required list size more than any change in messaging ever will.

Why high activity produces no meetings

Run the numbers above with a reachable share of 45% and then 85%, holding everything else constant. The required list size roughly halves. Now invert it: a team working a fixed list where only 45% is reachable is doing the same volume of work for roughly half the output, and every dashboard will report full activity compliance while it happens.

Four diagnoses, in the order they are worth checking:

1. Reachability. What share of sequenced contacts had a verified email or mobile? If nobody can answer, this is the problem.
2. Title accuracy. Are you reaching the right level? Correct company, wrong seniority produces polite silence rather than rejection, which is why it survives so long undetected.
3. Deliverability. High bounce rates damage sending reputation, which suppresses delivery to the valid addresses too. One bad list can degrade a domain for months.
4. Message. Genuinely last. It is the most-discussed variable and, in teams with a data problem, the least explanatory one.

What Prospecting Actually Costs

The two cost figures in Figure 5 are the same programme priced two ways, and the gap between them is the entire argument about data pricing models.

Under per-match pricing, you pay only for records that resolve. Your data cost per meeting is your unit cost divided by your response rate and your response-to-meeting rate. Reachability drops out of the equation entirely, because you were never charged for the records that failed.

Under per-record or per-credit pricing, you pay for every lookup whether it returns anything or not. Reachability stays in the denominator, so every point of coverage you lack shows up directly in your cost per meeting.

Putting data cost in proportion

Data is almost never the largest line in a prospecting programme. Fully loaded, an SDR typically costs multiples of the entire data budget for the list they work. That fact cuts in an unhelpful direction more often than teams realise.

Because data is the small number, it is the easy one to cut, and because its effects are lagged and indirect, the cut looks free for about a quarter. What actually happens is that the expensive resource, rep time, gets pointed at records that cannot be reached. Underspending on the cheap input to protect the budget wastes the costly one.

The number worth tracking is cost per reachable contact, not cost per record. It is the only figure that compares vendors on different pricing models, and the only one that reflects what you can actually do with what you bought.

Your Prospecting Plan: A Weekly Operating Cadence

Most prospecting programmes fail on consistency rather than technique. The plan below is deliberately boring. Its purpose is to make prospecting a scheduled operation with named owners rather than something that happens when the pipeline gets frightening.

Swipe the table sideways to see every column →

Prospecting skills, and how to standardise them across a team

Prospecting skill is usually described as resilience and persistence. Those matter, but they are not what separates teams that scale from teams that do not. Four capabilities are teachable, measurable and transferable:

• Research efficiency. Finding the one relevant fact about an account in three minutes rather than twenty. Cap the time box explicitly, because unbounded research is the most common form of productive-looking avoidance.
• Written concision. Getting a relevant, specific ask into under 90 words. Most cold email fails on length before it fails on content.
• Conversational control. Opening a call, earning thirty seconds, and asking a question that a busy person can answer without preparation.
• Data hygiene. Logging outcomes and reason codes accurately. Unglamorous, and the single behaviour that determines whether the team can learn anything from a quarter of work.

Standardising these across a team is a documentation problem, not a training problem. Write down the target personas, the qualifying questions, the objection responses and the disqualification criteria in one place, then have managers review recorded calls and sent emails against it. A prospecting playbook that lives in one experienced rep's head is a single point of failure with a notice period.

Metrics That Actually Matter

Most prospecting dashboards measure effort. Effort is not the constraint. These six measure the constraint.

Swipe the table sideways to see every column →

Compliance Guardrails

Most prospecting data is personal data, but not all of it, and the distinction is worth getting right. In the EU and UK, an email address that identifies a named individual, such as firstname.lastname@company.com, is personal data. So is a personal mobile number, and it carries more weight because it is more directly linked to the individual. A generic role address such as info@ or sales@ generally is not personal data, because it identifies a function rather than a person.

Where the data is personal, the obligations attach to you as the sender, not only to the vendor who supplied the record. Buying a list does not transfer responsibility.

• Lawful basis. Business-to-business outreach in the EU and UK is typically conducted under legitimate interests, which requires a documented balancing assessment. It is not a checkbox.
• Article 14 notice. Where personal data is obtained from a source other than the individual, you generally owe them a privacy notice, in practice at first contact.
• Do-not-call registers. Screening obligations differ by country. Assume any list containing mobile numbers needs screening against the relevant national register before dialling.
• Source provenance. Ask your provider where a record came from and on what basis it is processed. "We aggregate from public sources" is not an answer you can hand to a regulator.
• Suppression that actually works. Opt-outs must propagate back to the source list, not just the sequencer. Re-importing a suppressed contact next quarter is the most common self-inflicted breach.

We cover the European position in far more depth in Best European Email & Phone Enrichment Data Providers, including a country-by-country do-not-call reference and a template Article 14 notice.

Seven Ways Prospecting Fails

Inverting the question is more useful than another list of best practices. If you wanted a prospecting programme to fail while looking busy, here is how you would do it.

1. Buy volume before measuring reachability. Ten thousand records at 40% reachable is four thousand records and six thousand invoices.
2. Report activity, not reach. Count sends and dials rather than deliveries and connects, and you will never see the constraint.
3. Enrich once and never refresh. Contact data decays continuously; a static list is a depreciating asset with no depreciation schedule.
4. Sequence unreachable records anyway. They cost you the same rep capacity and pollute every conversion metric.
5. Let each rep build their own list. Unmeasurable, unrepeatable, and lost the day they leave.
6. Push deal qualification into the first touch. You are asking for budget from someone who has not agreed they have a problem.
7. Debate messaging when the data is broken. The most expensive meeting in B2B sales is the one where six people rewrite a subject line to fix a 41% bounce rate.

The Prospecting Tool Stack

Prospecting tools sell themselves as overlapping platforms, which makes comparison unnecessarily hard. They are easier to reason about as four distinct layers, each answering a different question.

Swipe the table sideways to see every column →

The most common mis-sequencing is buying layers 1, 3 and 4 while assuming layer 2 came free with layer 1. It rarely does. Broad platforms optimise for database breadth, which is a different problem from resolving a specific named individual to a working mobile number.

A second point worth being blunt about: layers 1 and 3 are frequently sold together as a seat licence, which means your cost scales with headcount rather than with usage. Layer 2 is the one place where per-result pricing is available, and it is worth taking, because it is the only layer where the vendor can fail to deliver and you can tell.

Where ZoomInfo, Apollo, Cognism and Lusha actually fit

The reason prospecting tool comparisons are so unsatisfying is that the products being compared are doing different jobs while using the same vocabulary. Almost everything in this market resolves to one of three jobs, and most confusion comes from assuming a tool that does one does all three.

ZoomInfo and Apollo are primarily platform plays with account intelligence attached. Cognism sits between platform and enrichment, with particular strength in European phone data. Lusha is closest to pure contact lookup. Clay is an orchestration layer that buys enrichment from others. TargetWise does contact enrichment and nothing else.

Two consequences follow, and they are the practical reason this framing matters:

• A single-vendor database has a fixed ceiling. If a broad platform does not hold a mobile number for your prospect, no amount of spending inside that platform produces one. The ceiling is architectural, not commercial. This is the whole argument for querying multiple sources rather than one.
• Seat pricing and result pricing measure different things. A platform licence charges for access whether or not the data resolves. Per-match pricing charges only when it does. Neither is inherently better, but they answer different questions, and only one of them lets you calculate cost per reachable contact.

Should you outsource prospecting?

Outsourced SDR agencies and prospecting services sell an appealing proposition: pipeline without hiring. They can work, but they fail predictably, and the failure mode is worth understanding before you sign.

An agency can supply three things: rep capacity, process, and speed to launch. It cannot supply the two things that most often constrain a programme. It cannot give you product knowledge deep enough to handle a technical objection in the first ninety seconds, and it cannot fix your data coverage, because it is buying from the same providers you are.

What that means in practice:

• Outsource when the constraint is capacity. You know the motion works, you have measured reachability, and you simply need more hours against the list.
• Do not outsource to discover the motion. If you have never booked a meeting from cold outbound, an agency is being paid to run an experiment you have not designed.
• Own the data either way. If the agency supplies the list, you are renting an asset and you will not keep it when the contract ends. Supply the list yourself, or contract for the data to be delivered to you.
• Contract on qualified meetings held, not meetings booked. Booked-meeting targets reliably produce no-shows.

How to Evaluate a B2B Prospecting Data Provider

Nine questions. The useful ones are the ones vendors do not lead with.

Swipe the table sideways to see every column →

Run the same 300 records through every provider on your shortlist and score the outputs yourself. It is the only comparison that is not marketing. Our head-to-head breakdowns of the major platforms are here: ZoomInfo, Apollo.io, Cognism, Lusha and RocketReach.

Key Takeaways

• Prospecting is the first step of the sales process, and it is a data operation with a conversation at the end of it, not a pure persuasion exercise.
• A prospect is someone you can actually contact. A name with no verified email or mobile is a hypothesis, and counting it distorts every ratio downstream.
• Reachability is the ceiling on everything else. Measure it before you spend on messaging, volume or intent data.
• Your data chooses your channel mix. A call-led motion is a purchasing decision about mobile coverage before it is a strategy.
• Buy the stack in order: targeting, then contact data, then engagement tooling, then signal. Most teams buy one, three and four and assume two was included.
• Work the list by fit and intent, not alphabetically, and remove the low-fit, low-intent segment from active campaigns entirely.
• Split your denominators. Response rate over contacts reached tells you about your messaging; response rate over contacts loaded tells you nothing you can act on.

Frequently Asked Questions

A Note on Method

Every figure in this article is a model you drive with your own inputs, not a benchmark. We have deliberately not published industry-wide coverage, bounce or reply rates, because those numbers vary so widely by country, seniority and segment that a single global figure is misleading more often than it is useful. Where a number appears in a chart, it is either arithmetic applied to values you set, or a clearly labelled worked example.

The one claim we will make without qualification is structural rather than statistical: prospecting output is bounded by the share of your list you can physically reach, and most teams have never measured that share. Measuring it costs one afternoon and a sample of 500 records.