Account based marketing (ABM): a practical guide for B2B startups
Account based marketing (ABM) is a B2B approach where sales and marketing agree on a named list of companies they want as customers, then point ads, outbound email, LinkedIn, and sales outreach at several people inside each one. You stop counting leads and start counting accounts: how many of the right companies know you, are talking to you, and have a deal open.
This guide covers how ABM works at a startup, how to choose the accounts, a six-week ABM campaign you can copy, and how to tell if it is working. If you want someone to build it, see the ABM agency page. For software, see the ABM platforms comparison.
Key takeaways
- ABM starts with the list. Score accounts for fit, then use buying signals to decide which ones get attention this month.
- Split the list into tiers. A few dozen accounts get one-to-one work, clusters of similar accounts get one-to-few, and the rest get signal-triggered one-to-many.
- Run ads, outbound email, LinkedIn, and the account owner against the same people in the same weeks. A six-week plan is below.
- Measure by account: coverage, engaged accounts, and pipeline per target account, compared with accounts outside the program.
- You do not need an ABM platform to start. A scored list, a CRM, and outbound tools are enough for the first campaign.
What ABM looks like in practice
Most B2B marketing works like a net. You publish, advertise, and run events, collect whoever fills in a form, and pass the good ones to sales. ABM works the other way round. Sales and marketing sit down, pick the companies that would be the best customers, and build every campaign around reaching the people at those companies who buy.
For a startup, that usually means a list of a few hundred to a few thousand accounts, a buying committee of three to six people at the important ones, and a small team covering several channels at once. ABM fits when deals are big enough that one logo matters, when a committee signs off rather than one person, and when your total market is small enough to name. If you sell a cheap tool that one user buys with a credit card, broader demand generation usually pays back faster.
One-to-one, one-to-few, one-to-many
Most ABM programs run three tiers at once. What changes between them is how much custom work each account gets.
| Tier | Typical size at a startup | Work per account | Main channels |
|---|---|---|---|
| One-to-one | 10 to 50 accounts | Hours of research, messaging written for that company, executive involvement | Founder or AE outreach, personal video, direct mail, a dinner or event invite |
| One-to-few | 100 to 500 accounts in clusters of similar companies | Messaging per cluster (an industry, a use case, a competitor they use) | Outbound by role, LinkedIn ads per cluster, a proof asset per cluster |
| One-to-many | 500 to a few thousand accounts | Little until a signal fires, then a templated play | Signal-triggered outbound, broad matched audiences, retargeting |
Accounts move between tiers. A one-to-many account where three people visit the pricing page in a week deserves one-to-few treatment until it goes quiet again.
How to pick target accounts
Target account selection decides most of the result. Use two inputs: fit, which says whether a company could be a good customer, and signals, which say whether now is a good time.
- Write fit as filters. Industry, employee band, region, tech stack, funding stage, and anything else a database can check. Then count the accounts that pass. The TAM calculator helps size it.
- Score fit from your won deals. Look at your best customers and weight the traits they share. A company that matches four of five traits ranks above one that matches two.
- Layer in signals. Hiring for a role your product supports, a new leader in the buying seat, fresh funding, a tech stack change, or repeat visits to your site. The buying signals guide lists the ones worth tracking, and intent data providers covers third-party intent.
- Sort into tiers. High fit plus a strong signal goes to one-to-one or one-to-few. High fit with no signal waits in one-to-many until something happens.
- Get sales to sign off. Every account owner reviews their list and can swap out a few. Then freeze it for the quarter so the numbers mean something.
- Map the buying committee. For tier one and two accounts, find and verify the economic buyer, the likely champion, a user or two, and anyone who could block the deal.
A sample 6-week ABM campaign plan
Here is a campaign for about 200 one-to-few accounts with a small team: a marketer, an SDR or two, the AEs who own the accounts, and whoever runs ads. Swap the owners to match your team.
| Week | Channel | Action | Owner |
|---|---|---|---|
| 1 | Data | Score and tier the accounts, attach the current signal to each, and get sign-off from account owners | RevOps or GTM engineer |
| 1 | Data | Find and verify three to six contacts per account, and load them to the CRM with tier and cluster | RevOps or GTM engineer |
| 2 | Ads | Upload the company list and contact list to LinkedIn and start awareness ads to the committee | Paid team |
| 2 | Write copy per role and cluster, and warm any new sending domains | Marketing and SDR | |
| 3 | First outbound touch to champions and users, opening with the account's signal | SDR | |
| 3 | Connection requests from the AE to economic buyers, with no pitch in the note | AE | |
| 4 | Email and LinkedIn | Second touch with a proof asset: a teardown, a benchmark, or a short customer story from their industry | SDR |
| 4 | Ads | Switch creative to the same proof asset for accounts that engaged | Paid team |
| 5 | Sales | Personal outreach from the account owner to every account with two or more engaged people | AE |
| 5 | Direct mail or event | Optional: a gift or dinner invite for the handful of accounts closest to a meeting | Marketing |
| 6 | Review | Compare engaged accounts, meetings, and pipeline by cluster, against accounts outside the campaign. Decide what to keep for the next round. | Marketing lead and sales lead |
Two practical notes. LinkedIn will only serve an ad set to a list that matches at least 300 members, so very small one-to-one lists usually need to be combined with others or reached through outbound instead (LinkedIn's company list rules). And if your domains are new, start warming them in week one, because a cold domain sending to your best accounts is the fastest way to waste them. The email deliverability guide covers the setup.
Account based marketing examples
A security startup selling to fintechs (one-to-few)
The list: 150 fintech companies of 50 to 500 employees that posted a first security or compliance hire in the last 60 days. The committee: the CTO, the head of engineering, and the hiring manager. LinkedIn ads run to all three while the SDR emails the head of engineering about the gap a new security hire is usually brought in to close. The AE calls any account where two or more people engage. The hiring post gives every message a reason to exist.
Vertical software chasing 25 enterprise logos (one-to-one)
A logistics software company names the 25 largest shippers in its region. For each one, someone reads the annual report and recent news, finds the operations and finance leaders, and writes a one-page note on where that company is losing money in their process. The founder sends it personally, with a follow-up from the AE and an invite to a small operator dinner. Ads play no part, because the list is too small for them to serve.
A data tool working 2,000 accounts by signal (one-to-many)
A data infrastructure startup scores 2,000 companies that use a specific warehouse. All of them see broad LinkedIn ads. Outbound only starts when a signal fires: a job post mentioning the warehouse, a new head of data, or a repeat visit to the docs. Each signal has its own short sequence, and an account that engages in two channels moves into a one-to-few cluster for the rest of the quarter.
How to measure an ABM program
ABM fails lead-based reporting by design, since a program aimed at 200 accounts will never produce thousands of form fills. Measure by account, by tier, and against a control group of similar accounts you left out.
| Metric | How to count it | What it tells you |
|---|---|---|
| Coverage | Share of target accounts with at least three verified contacts in the CRM | Whether the program can reach the committee at all |
| Account engagement | Share of target accounts where two or more people replied, visited key pages, engaged with ads, or took a meeting | Whether the accounts are paying attention |
| Meetings per tier | First meetings held, by tier and cluster | Which clusters and messages are working |
| Pipeline per target account | Pipeline created in the tier divided by accounts in the tier | The headline number, comparable across tiers and quarters |
| Win rate and deal size | Against accounts outside the program over the same period | Whether ABM changes outcomes, or just takes credit for deals that would have happened |
| Sales cycle | Days from first meeting to close, target accounts versus the rest | Whether a warmed committee moves faster |
Review every two weeks during a campaign, and judge pipeline over a full sales cycle. A six-week campaign at a company with a four-month cycle will show engagement and meetings long before it shows revenue.
Common ABM mistakes
- A target list marketing built alone. If the account owners did not agree to it, they will not work it.
- Picking accounts on fit only. Two thousand perfect-fit companies with no reason to talk this quarter is a long, slow list.
- Reaching one person per account. A committee buys, so a single contact who goes quiet ends the account.
- Running ads and outbound in different months, or to different lists, so neither warms the other.
- Buying an ABM platform before the first campaign. The software assumes a list, a process, and someone to run it.
- Reporting clicks and impressions. Executives care about meetings and pipeline per account.
Do you need an ABM platform?
Not for the first campaign. A scored list in your CRM, a contact data source, matched audiences on LinkedIn, and an outbound tool cover everything in the six-week plan above. Platforms earn their price once the program covers thousands of accounts, several teams share it, and someone owns it full time. The ABM platforms comparison covers the suites, the account-based ad tools, and what they cost, and website visitor identification covers the tools that tell you which accounts are on your site.
How OneGTM runs signal-based ABM
I build the part of ABM that sits under the ads. That means the scored and tiered account list, the buying committee at each account, signals wired to the list with alerts to the account owner, matched audiences pushed to LinkedIn and Google, and multi-threaded outbound by email and LinkedIn, timed to the ads. OneGTM does not run the ads; your paid team or agency runs them against the audiences I build. Everything lives in your CRM and tools, so it keeps running after handover.
For Patlytics, an AI patent platform, that meant intent signal aggregation and BDR enablement alongside the inbound and outbound build. The full scope is on the ABM agency page, and the GTM strategy guide covers the decisions that come before ABM.
Account based marketing FAQ
What is account based marketing?
What does ABM mean in marketing?
What is an ABM strategy?
What is an ABM campaign?
What is an example of account based marketing?
How do you measure an ABM program?
Is account based marketing worth it for a B2B startup?
Want an ABM program running on your own data, without buying a platform first?
Talk to OneGTM about account based marketing
Tell me what you're building. I reply the same day.
Pick a time
Nothing works? Write team@getaob.com and I'll find one.
Book a timeGrid not loading? Open the booking page
Rather do it the old way? team@getaob.com LinkedIn