Haku
Book a demo
← Blog
Prospecting · Sales Intelligence · Practical Guide

How to build a B2B prospect list that actually converts.

A list of 5,000 loosely matched contacts will produce fewer replies than a list of 500 tightly targeted ones. The goal is not a big list. It is a precise list of companies where something is happening right now that makes your outreach relevant — and contacts at those companies whose details will not bounce.

Cameron Evans
Cameron Evans · Founder & CEO, Haku
8 May 2026 · 10 min read

You have probably had this experience. You export a list from a data provider, load it into your sequencer, and watch the reply rate come back at under 2%. You blame the copy. You rewrite the subject line. You try a different sequence structure. The reply rate moves from 1.8% to 2.1% and you tell yourself you are making progress.

The copy is rarely the problem. The list almost always is.

A prospect list built on imprecise ICP criteria, with no signal filtering and unvalidated contact data, will underperform regardless of how well the sequences are written. The difference between a 2% reply rate and a 10% reply rate is almost always data quality and targeting precision — not messaging. This guide covers how to build a prospect list correctly from the ground up.

What a high-quality prospect list actually is

Before building anything, it is worth being precise about what you are building. A prospect list is not a contact list. A contact list is everyone who might theoretically buy from you. A prospect list is the subset of those people who match your ICP precisely, are likely to be in an active buying window, and have verified, reachable contact information.

A high-quality prospect list has three characteristics — and most teams only build for one of them.

Fit — the companies and contacts match your ICP criteria precisely: industry, company size, geography, tech stack, funding stage, role, and seniority. Not broadly, but specifically enough that the challenge your product solves is genuinely relevant to this person at this company right now.

Freshness — the contact data has been validated recently, not six months ago when the database was last refreshed. B2B contact data decays at 25 to 30% annually. A contact who was at the right company when the database was updated may have moved on by the time your sequence lands.

Signal context — the accounts have been filtered by buying signals that indicate an active purchasing window, not just demographic filters that indicate they might buy eventually. A company that raised a funding round last week, hired a new VP of Sales, or is actively growing its SDR team is in a fundamentally different position to one that simply matches your firmographic criteria.

× The wrong approach
Export 5,000 contacts from a database filtered by industry and headcount. Load into sequencer. Send. Wonder why reply rate is 1.8%. Conclude that outbound doesn't work.
✓ The right approach
Build 200–500 contacts filtered by ICP + buying signals + verified data. Every account has a specific reason to receive outreach today. Reply rate 6–10%. Outbound works.

The six-step framework for building a prospect list that converts

1
Define your ICP from evidence, not assumptions
Go back through your closed-won deals from the last 12 to 24 months. Look for patterns that go beyond firmographics — what did these companies have in common at the moment they bought? What triggered the conversation? What was the business context? What roles were involved in the decision? Your ICP is not a demographic description of who might buy. It is an evidence-based model of who has bought, and why.
2
Build your account list — companies before contacts
Always start at the company level, not the contact level. Identify the accounts that fit your ICP before finding the individual contacts within them. This prevents the common mistake of finding a contact at the wrong company and working backwards to justify the inclusion. Filter by industry, headcount, revenue range, geography, tech stack, and funding stage. Then layer in signal filters to identify which of those accounts are in an active buying window right now.
3
Layer in buying signals to prioritise
A list of ICP-matched accounts tells you who might buy. A list filtered by buying signals tells you who is most likely buying right now. For each account on your company list, look for the signals that indicate an active purchasing window — a recent funding event, a new senior hire, headcount growth in the relevant function, a technology change, a new market launch. Accounts with two or more aligned signals go to the top of your list. Accounts with no signals are deprioritised, even if they match your ICP perfectly on paper.
4
Identify the right contacts within each account
Once your account list is built and prioritised, find the right decision-maker or influencer within each company. The right contact depends on what you sell — it is not always the most senior person. Map the buying committee: who makes the decision, who influences it, who will use the product, and who controls the budget. For most mid-market outbound, a single contact per account is a starting point; multi-threading across two or three stakeholders at your highest-priority accounts produces better results.
5
Enrich and validate every contact before it reaches a sequence
Source contact details — email address and phone number — through multi-provider enrichment rather than a single static database. Validate every email address against live SMTP checks at the point of use. Never load a contact into a sequence until you have confirmed the email address is active and deliverable. A 2% or lower bounce rate is achievable. Above 3% means the data is already degraded and every send is damaging your sender domain.
6
Segment by signal for outreach that feels personal
Do not load all 500 contacts into a single generic sequence. Segment by signal type — funding event contacts get a different first message to new-hire contacts, which gets a different message to tech-change contacts. The segmentation does not need to be elaborate. Two or three segments, each with a first touch that references the relevant business context, produces meaningfully better engagement than one generic sequence sent to everyone.

Building the ICP filter — what to include

The ICP filter is the specification document for your prospect list. Every sourcing decision should be evaluated against it. The more specific it is, the better your list will perform — a well-defined ICP should disqualify at least 70% of the accounts that superficially look like a fit.

Firmographic filters
Industry / vertical · Company headcount range · Annual revenue · Geography / market · Business model (B2B, B2C, marketplace) · Funding stage
Technographic filters
CRM in use (Salesforce, HubSpot) · Sales engagement platform · Marketing automation stack · Key integrations relevant to your product
Contact filters
Job title / function · Seniority level · Department (Sales, RevOps, Marketing) · Time in role · Relevant background or experience
Behavioural filters
Recent funding round · Headcount growth in target function · New leadership hire · CRM or tech stack migration · Job postings for relevant roles
The negative ICP matters as much as the positive one.

Defining which accounts you will deliberately not pursue is as important as defining which ones you will. Companies below a certain headcount, in the wrong geography, using a competing product that would require displacement — all of these should be explicitly excluded. Disqualifying accounts early is far less expensive than running them through a sequence and discovering the mismatch in month two.

Signal prioritisation — which accounts to call this week

Once your ICP filter is defined, signals determine which accounts within that filter to prioritise for outreach this week versus next month versus never. Not all signals are equal — the strongest indicators of an active buying window are the ones that reflect a genuine organisational change, not passive interest.

Signal
Priority
Why it matters
Outreach window
Funding round + new VP Sales hire (combined)
Highest
Budget available + stack under active review. The strongest combined signal in B2B outbound.
48–72 hours
New VP Sales / CRO / RevOps hire
High
New leaders audit and replace their inherited stack in the first 90 days. Window is narrow and competitive.
48–72 hours
Funding round (standalone)
High
Budget is moving. Growth is the mandate. Investment in tooling follows capital raise consistently.
48–72 hours
SDR / AE headcount growth (3+ hires in 60 days)
Medium-high
Scaling sales motion requires the infrastructure to support it. Timing is good, urgency is building.
1 week
CRM or tech stack migration
Medium
Active re-evaluation of the entire stack. Everything connected to the CRM is up for review.
1 week
New product launch / market expansion
Medium
New ICP means the existing data and tools may not serve the new motion. Creates a genuine need.
1–2 weeks
ICP match only — no active signal
Low
Could buy eventually. No evidence of an active window. Deprioritise until a signal fires.
Monitor — do not contact yet

Data sourcing and enrichment — the step most teams skip

Finding an account that matches your ICP and has the right signals is step one. Step two is sourcing verified contact data for the decision-maker within that account — and this is where most prospect lists degrade before they are ever used.

The instinct is to export contacts from a single data provider, import into a sequencer, and start sending. The problem is that single-source databases — even the best ones — have fill rates of 55 to 65% on email addresses for mid-market accounts, and bounce rates of 8 to 25% on the emails they do return. By the time a campaign starts, a significant proportion of the contacts are unreachable and the sends to stale addresses are damaging the sending domain for every future campaign.

The correct approach is waterfall enrichment: query multiple data providers in sequence for each contact, moving to the next provider only when the previous one fails to return a verified result. Then validate the returned email against a live SMTP check — not a format validation, but a check that confirms the mailbox is active and accepting mail right now, not at the moment the database was last refreshed.

55–65%
Email fill rate from single-source databases for mid-market accounts — 35–45% return nothing
80%+
Email match rate achievable through waterfall enrichment across 15+ providers
<2%
Target bounce rate for a well-built prospect list — achievable with verified, multi-source contact data

List hygiene — the ongoing maintenance most teams ignore

A prospect list is not a one-time deliverable. It degrades continuously. The contacts you sourced three months ago have a 7.5% probability of being inaccurate today — and that number climbs every week you leave the list unrefreshed.

Standard list hygiene practice is to re-verify email addresses for any contact who has not been sequenced within six months, to remove hard bounces from the active list immediately rather than allowing them to accumulate, and to update job titles and company affiliations for contacts in high-turnover sectors — technology, finance, growth-stage companies — where role changes happen more frequently than in stable industries.

Remove bounced contacts immediately.

Every hard bounce signals to Gmail and Outlook that you are sending to non-existent addresses. Those signals accumulate and degrade your sender domain reputation — not just for the bounced contact, but for every email that follows from that domain. A bounce rate that climbs above 3% needs to be fixed at the data layer, not the deliverability layer.

Most small B2B teams need 200 to 500 fresh, verified contacts per month to sustain a healthy outbound pipeline. That means list building is not a quarterly exercise — it is a continuous workflow that runs alongside the outreach motion, producing a consistent flow of new, signal-qualified, verified contacts to replace those that have been sequenced, have bounced, or have exited the buying window.

How Haku builds the list for you

Every step in this framework represents time and skill that currently sits in a rep's workflow. Defining the ICP from CRM data, identifying accounts, sourcing buying signals, finding decision-maker contacts, enriching and validating, segmenting by signal — that is the research layer of outbound, and it is consuming the majority of time most reps spend on prospecting every week.

Haku was built to automate every part of it. Before surfacing a single account, Haku ingests your CRM data, your closed-won history, your case studies, and your product information to build a custom ICP model specific to your business. It then monitors every relevant public data source — LinkedIn, job boards, Companies House, Google Maps, news, forums — for the signals that indicate a buying window, and surfaces accounts when the right combination fires.

Contact data is validated through FullEnrich, aggregating 17+ independent providers, with triple-layer SMTP verification before any contact reaches a rep. The output is a pre-built, signal-qualified, validated prospect list — ready to sequence against, segmented by signal type, with bounce rates consistently below 3%.

A rep using Haku starts their day with their list already built. They spend their time on the conversation — not the six steps that preceded it.

References
  1. Salesforce. (2026). State of Sales Report, 7th Edition. Salesforce Research. salesforce.com
  2. Gartner. (2025). Future of Sales: Buyer Enablement and Buying Group Dynamics. Gartner Research. gartner.com
  3. Forrester Research. (2026). State of Business Buying 2026. Forrester. forrester.com
  4. LinkedIn Sales Solutions. (2026). State of Sales Report. LinkedIn. linkedin.com
  5. FullEnrich. (2026). Waterfall Enrichment: How It Works. FullEnrich. fullenrich.com

See Haku build a prospect list from your own ICP.

Book a demo
How to build a B2B prospect list that actually converts. — Haku
← Blog
Prospecting · Sales Intelligence · Practical Guide

How to build a B2B prospect list that actually converts.

A list of 5,000 loosely matched contacts will produce fewer replies than a list of 500 tightly targeted ones. The goal is not a big list. It is a precise list of companies where something is happening right now that makes your outreach relevant — and contacts at those companies whose details will not bounce.

Cameron Evans
Cameron Evans · Founder & CEO, Haku
8 May 2026 · 10 min read

You have probably had this experience. You export a list from a data provider, load it into your sequencer, and watch the reply rate come back at under 2%. You blame the copy. You rewrite the subject line. You try a different sequence structure. The reply rate moves from 1.8% to 2.1% and you tell yourself you are making progress.

The copy is rarely the problem. The list almost always is.

A prospect list built on imprecise ICP criteria, with no signal filtering and unvalidated contact data, will underperform regardless of how well the sequences are written. The difference between a 2% reply rate and a 10% reply rate is almost always data quality and targeting precision — not messaging. This guide covers how to build a prospect list correctly from the ground up.

What a high-quality prospect list actually is

Before building anything, it is worth being precise about what you are building. A prospect list is not a contact list. A contact list is everyone who might theoretically buy from you. A prospect list is the subset of those people who match your ICP precisely, are likely to be in an active buying window, and have verified, reachable contact information.

A high-quality prospect list has three characteristics — and most teams only build for one of them.

Fit — the companies and contacts match your ICP criteria precisely: industry, company size, geography, tech stack, funding stage, role, and seniority. Not broadly, but specifically enough that the challenge your product solves is genuinely relevant to this person at this company right now.

Freshness — the contact data has been validated recently, not six months ago when the database was last refreshed. B2B contact data decays at 25 to 30% annually. A contact who was at the right company when the database was updated may have moved on by the time your sequence lands.

Signal context — the accounts have been filtered by buying signals that indicate an active purchasing window, not just demographic filters that indicate they might buy eventually. A company that raised a funding round last week, hired a new VP of Sales, or is actively growing its SDR team is in a fundamentally different position to one that simply matches your firmographic criteria.

× The wrong approach
Export 5,000 contacts from a database filtered by industry and headcount. Load into sequencer. Send. Wonder why reply rate is 1.8%. Conclude that outbound doesn't work.
✓ The right approach
Build 200–500 contacts filtered by ICP + buying signals + verified data. Every account has a specific reason to receive outreach today. Reply rate 6–10%. Outbound works.

The six-step framework for building a prospect list that converts

1
Define your ICP from evidence, not assumptions
Go back through your closed-won deals from the last 12 to 24 months. Look for patterns that go beyond firmographics — what did these companies have in common at the moment they bought? What triggered the conversation? What was the business context? What roles were involved in the decision? Your ICP is not a demographic description of who might buy. It is an evidence-based model of who has bought, and why.
2
Build your account list — companies before contacts
Always start at the company level, not the contact level. Identify the accounts that fit your ICP before finding the individual contacts within them. This prevents the common mistake of finding a contact at the wrong company and working backwards to justify the inclusion. Filter by industry, headcount, revenue range, geography, tech stack, and funding stage. Then layer in signal filters to identify which of those accounts are in an active buying window right now.
3
Layer in buying signals to prioritise
A list of ICP-matched accounts tells you who might buy. A list filtered by buying signals tells you who is most likely buying right now. For each account on your company list, look for the signals that indicate an active purchasing window — a recent funding event, a new senior hire, headcount growth in the relevant function, a technology change, a new market launch. Accounts with two or more aligned signals go to the top of your list. Accounts with no signals are deprioritised, even if they match your ICP perfectly on paper.
4
Identify the right contacts within each account
Once your account list is built and prioritised, find the right decision-maker or influencer within each company. The right contact depends on what you sell — it is not always the most senior person. Map the buying committee: who makes the decision, who influences it, who will use the product, and who controls the budget. For most mid-market outbound, a single contact per account is a starting point; multi-threading across two or three stakeholders at your highest-priority accounts produces better results.
5
Enrich and validate every contact before it reaches a sequence
Source contact details — email address and phone number — through multi-provider enrichment rather than a single static database. Validate every email address against live SMTP checks at the point of use. Never load a contact into a sequence until you have confirmed the email address is active and deliverable. A 2% or lower bounce rate is achievable. Above 3% means the data is already degraded and every send is damaging your sender domain.
6
Segment by signal for outreach that feels personal
Do not load all 500 contacts into a single generic sequence. Segment by signal type — funding event contacts get a different first message to new-hire contacts, which gets a different message to tech-change contacts. The segmentation does not need to be elaborate. Two or three segments, each with a first touch that references the relevant business context, produces meaningfully better engagement than one generic sequence sent to everyone.

Building the ICP filter — what to include

The ICP filter is the specification document for your prospect list. Every sourcing decision should be evaluated against it. The more specific it is, the better your list will perform — a well-defined ICP should disqualify at least 70% of the accounts that superficially look like a fit.

Firmographic filters
Industry / vertical · Company headcount range · Annual revenue · Geography / market · Business model (B2B, B2C, marketplace) · Funding stage
Technographic filters
CRM in use (Salesforce, HubSpot) · Sales engagement platform · Marketing automation stack · Key integrations relevant to your product
Contact filters
Job title / function · Seniority level · Department (Sales, RevOps, Marketing) · Time in role · Relevant background or experience
Behavioural filters
Recent funding round · Headcount growth in target function · New leadership hire · CRM or tech stack migration · Job postings for relevant roles
The negative ICP matters as much as the positive one.

Defining which accounts you will deliberately not pursue is as important as defining which ones you will. Companies below a certain headcount, in the wrong geography, using a competing product that would require displacement — all of these should be explicitly excluded. Disqualifying accounts early is far less expensive than running them through a sequence and discovering the mismatch in month two.

Signal prioritisation — which accounts to call this week

Once your ICP filter is defined, signals determine which accounts within that filter to prioritise for outreach this week versus next month versus never. Not all signals are equal — the strongest indicators of an active buying window are the ones that reflect a genuine organisational change, not passive interest.

Signal
Priority
Why it matters
Outreach window
Funding round + new VP Sales hire (combined)
Highest
Budget available + stack under active review. The strongest combined signal in B2B outbound.
48–72 hours
New VP Sales / CRO / RevOps hire
High
New leaders audit and replace their inherited stack in the first 90 days. Window is narrow and competitive.
48–72 hours
Funding round (standalone)
High
Budget is moving. Growth is the mandate. Investment in tooling follows capital raise consistently.
48–72 hours
SDR / AE headcount growth (3+ hires in 60 days)
Medium-high
Scaling sales motion requires the infrastructure to support it. Timing is good, urgency is building.
1 week
CRM or tech stack migration
Medium
Active re-evaluation of the entire stack. Everything connected to the CRM is up for review.
1 week
New product launch / market expansion
Medium
New ICP means the existing data and tools may not serve the new motion. Creates a genuine need.
1–2 weeks
ICP match only — no active signal
Low
Could buy eventually. No evidence of an active window. Deprioritise until a signal fires.
Monitor — do not contact yet

Data sourcing and enrichment — the step most teams skip

Finding an account that matches your ICP and has the right signals is step one. Step two is sourcing verified contact data for the decision-maker within that account — and this is where most prospect lists degrade before they are ever used.

The instinct is to export contacts from a single data provider, import into a sequencer, and start sending. The problem is that single-source databases — even the best ones — have fill rates of 55 to 65% on email addresses for mid-market accounts, and bounce rates of 8 to 25% on the emails they do return. By the time a campaign starts, a significant proportion of the contacts are unreachable and the sends to stale addresses are damaging the sending domain for every future campaign.

The correct approach is waterfall enrichment: query multiple data providers in sequence for each contact, moving to the next provider only when the previous one fails to return a verified result. Then validate the returned email against a live SMTP check — not a format validation, but a check that confirms the mailbox is active and accepting mail right now, not at the moment the database was last refreshed.

55–65%
Email fill rate from single-source databases for mid-market accounts — 35–45% return nothing
80%+
Email match rate achievable through waterfall enrichment across 15+ providers
<2%
Target bounce rate for a well-built prospect list — achievable with verified, multi-source contact data

List hygiene — the ongoing maintenance most teams ignore

A prospect list is not a one-time deliverable. It degrades continuously. The contacts you sourced three months ago have a 7.5% probability of being inaccurate today — and that number climbs every week you leave the list unrefreshed.

Standard list hygiene practice is to re-verify email addresses for any contact who has not been sequenced within six months, to remove hard bounces from the active list immediately rather than allowing them to accumulate, and to update job titles and company affiliations for contacts in high-turnover sectors — technology, finance, growth-stage companies — where role changes happen more frequently than in stable industries.

Remove bounced contacts immediately.

Every hard bounce signals to Gmail and Outlook that you are sending to non-existent addresses. Those signals accumulate and degrade your sender domain reputation — not just for the bounced contact, but for every email that follows from that domain. A bounce rate that climbs above 3% needs to be fixed at the data layer, not the deliverability layer.

Most small B2B teams need 200 to 500 fresh, verified contacts per month to sustain a healthy outbound pipeline. That means list building is not a quarterly exercise — it is a continuous workflow that runs alongside the outreach motion, producing a consistent flow of new, signal-qualified, verified contacts to replace those that have been sequenced, have bounced, or have exited the buying window.

How Haku builds the list for you

Every step in this framework represents time and skill that currently sits in a rep's workflow. Defining the ICP from CRM data, identifying accounts, sourcing buying signals, finding decision-maker contacts, enriching and validating, segmenting by signal — that is the research layer of outbound, and it is consuming the majority of time most reps spend on prospecting every week.

Haku was built to automate every part of it. Before surfacing a single account, Haku ingests your CRM data, your closed-won history, your case studies, and your product information to build a custom ICP model specific to your business. It then monitors every relevant public data source — LinkedIn, job boards, Companies House, Google Maps, news, forums — for the signals that indicate a buying window, and surfaces accounts when the right combination fires.

Contact data is validated through FullEnrich, aggregating 17+ independent providers, with triple-layer SMTP verification before any contact reaches a rep. The output is a pre-built, signal-qualified, validated prospect list — ready to sequence against, segmented by signal type, with bounce rates consistently below 3%.

A rep using Haku starts their day with their list already built. They spend their time on the conversation — not the six steps that preceded it.

References
  1. Salesforce. (2026). State of Sales Report, 7th Edition. Salesforce Research. salesforce.com
  2. Gartner. (2025). Future of Sales: Buyer Enablement and Buying Group Dynamics. Gartner Research. gartner.com
  3. Forrester Research. (2026). State of Business Buying 2026. Forrester. forrester.com
  4. LinkedIn Sales Solutions. (2026). State of Sales Report. LinkedIn. linkedin.com
  5. FullEnrich. (2026). Waterfall Enrichment: How It Works. FullEnrich. fullenrich.com

See Haku build a prospect list from your own ICP.

Book a demo