Account-Based Marketing is a way of working in which marketing and sales pick a named list of companies and work them together, instead of catching whatever leads the market happens to produce. The unit of account is the company, not the lead.
The biggest change is organisational, not technical: one shared goal, one account list, one meeting where both sides look at the same numbers. Most failed ABM programmes fail there, not in the tooling.
ABM is not better than demand generation. It answers a different situation.
It fits when the contract is worth enough that one extra win changes the year, when several people have to agree before anyone signs, and when you already know roughly who your best customers are. At a 10,000 contract with a two-week cycle, ABM wastes money.
Three situations where I tell founders not to start. The product has no repeatable customer yet, so there is nothing to build a list from. Average contract value is below 30,000 a year, so the maths does not close. Or the leadership team does not agree on what a good customer is, which is an ownership decision no agency can settle.
The comparison with the alternative model is in ABM vs lead generation.
The order matters more than any individual step. Most programmes go wrong by starting in the middle, usually by buying a tool.
1. Understand why people actually bought. Twelve conversations of forty minutes with customers who stayed, customers who left, and companies that did not buy. This produces an ICP based on situation rather than industry, the language customers use, and the list of events that precede a purchase. See B2B customer interviews.
2. Map how they buy. Your CRM stages describe what your rep does. They say nothing about what is happening at the customer. Mapping the buying process is what turns a forecast from instinct into something you can question. See mapping the B2B buying and sales process.
3. Build and score the list. Fit tells you whether a company matches the profile. Heat tells you whether it is moving now. Two separate numbers, recalculated on different schedules, feeding a tiering model where the tiers actually differ in how much work an account gets. See account scoring in ABM.
4. Watch for signals. Hiring for a role that owns your problem, a stack change, a new department head, a third pricing visit in a week. Each has a window, and a signal outside its window is noise. See B2B buying signals and, for the bought kind, B2B intent data.
5. Run it as one team. One list visible to both sides in the CRM, one weekly review, one agreed definition of what gets handed over and how fast it is handled. Without that, marketing and sales work on two different sets of companies and meet only to argue.
A first programme runs on a CRM, a spreadsheet, a contact database and LinkedIn Ads. No platform required, and platforms rarely earn their place below a thousand accounts and three reps. See ABM without a platform.
On the money: implementation is 30,000 to 60,000 zloty net for the first three to four months, plus 2,000 to 6,000 a month on tools. The full breakdown and what drives the price up is in what ABM costs, and you can test it on your own figures in the ABM ROI calculator.
Whether to hire or bring in help is a separate question with an honest answer on both sides: see ABM agency or in-house team.
Pipeline first, revenue later. Pipeline forms faster than revenue and is harder to inflate than lead volume, which is why it is the metric I hold programmes to.
Alongside it: how many accounts from the list entered a conversation, how many have more than one person engaged, and win rate. Win rate usually improves before opportunity count does, because you are working better-fitting companies.
What not to measure in the first quarter: revenue. If your cycle is six months, the first win lands in month eight or nine. Judging the programme at three months is guessing.
And a warning about attribution. Last-click attribution almost always lies in ABM, because the process runs for months and touches several people. I have seen companies increase budget for a year on a channel that was only closing contacts generated elsewhere.
Paying marketing for MQL volume. If the compensation plan counts MQLs, you get MQLs. Not fit, not pipeline. Exactly what you paid for.
An account list with no owner and no review date. After a quarter nobody knows whether it is current, and it quietly stops being used.
Buying a platform before knowing which signals matter. The tool produces hundreds of alerts, nobody knows which to act on, and everyone mutes it within a quarter.
Working a single contact at each account. One person leaving kills the deal. Multithreading is the cheapest insurance in B2B sales and most reps still skip it.
Three things that need no budget. Pull your longest-staying, highest-paying customers out of the CRM and find what they share. Write the one-page agreement between sales and marketing and have both sign it. Start logging signals in a spreadsheet.
After six weeks of that you will know more about your market than any bought dataset would have told you, and you will know what is worth automating.
Every term in this guide, from ICP and target account list through MQA, intent data and NRR, is defined in the ABM glossary, which covers over a hundred entries in plain language.
Daniel Tupczyński runs MOAT Marketing, an Account-Based Marketing partner for B2B SaaS companies. MOAT builds and runs ABM systems alongside the client’s own sales and marketing team, from market research and prospecting into Tier 1 accounts through to insights for product and strategy, so the know-how stays in the company. He also runs marketing at a B2B SaaS company.