Account-based marketing in B2B: the work that happens before the campaign

Everyone wants to talk about orchestration and intent data. The programmes that actually return pipeline are built on three foundations nobody enjoys talking about: sharp positioning, a validated ICP, and a clean list. Get those right and the campaign almost runs itself. Get them wrong and no amount of budget will save it.

Account-based marketing has a marketing problem of its own. It gets sold as a technology decision. Buy the platform, upload the target list, switch on the ads, and the meetings arrive. The reality is quieter and a good deal more demanding. The accounts that convert are usually won or lost long before the first impression serves, in the work most teams skip because it does not look like much on a status update.

At Spanb2b we think about all of this through the gap between marketing and revenue. ABM sits in what we call the Account gap: the space where a named set of companies either turns into pipeline or turns into a spreadsheet nobody opens again. Closing that gap has very little to do with tooling and almost everything to do with three things done properly, and done in the right order.

Positioning comes before targeting

You cannot personalise a message you have not decided on. This sounds obvious, and yet it is the step teams are quickest to wave through. They jump straight to "which accounts" without settling "why us, why now, why not the incumbent." The result is technically personalised outreach that says nothing. The company name is right at the top of the email and the argument underneath it is generic.

Positioning is the piece that gives an ABM programme its spine. It forces the decisions that everything downstream depends on: the specific problem you are the obvious answer to, the category you want to be measured in, the alternatives you are displacing, and the proof that makes the claim credible to a sceptical buying committee. Until those are locked, tiering and channel planning are premature. You are optimising the delivery of a message that has not earned the right to land.

A useful test: if you swapped your logo for a competitor's, would the campaign still make sense? If the answer is yes, you have a targeting plan, not a positioning-led programme. The whole point of ABM is that the message is unmistakably yours and pointedly relevant to them. That only happens when positioning does its job first.

Your ICP is a hypothesis, not a wishlist

The ideal customer profile is where good intentions turn into fiction. Most ICPs are assembled in a workshop from a mix of aspiration and the accounts the sales team wishes it had. They describe the customer everyone would like, rather than the customer who actually buys, stays, and expands.

Validating the ICP means treating it as a claim to be tested against evidence. Look at who has actually closed, and how fast. Look at deal size, sales cycle length, win rate, and retention by segment, not just the logos on the case study wall. Interview the accounts that converted well and, more instructively, the ones that looked perfect on paper and went nowhere. Patterns emerge that no workshop predicts. The firmographics that correlate with revenue are frequently narrower and stranger than the ones on the original slide.

The discipline here is to follow the data wherever it leads, rather than confirming the account you were already hoping for, and to be willing to be wrong about who your best-fit customer actually is. A validated ICP is usually smaller than the one you started with, and that is the point. A tighter definition means every pound of budget lands on accounts with a genuine reason to buy, rather than being spread thinly across a list built to feel ambitious.

The list is the programme

Once positioning is sharp and the ICP is evidence-backed, the target list becomes the single most important asset in the programme. It is also the one most likely to be rotten underneath.

List cleansing is unglamorous and non-negotiable. Duplicate accounts, dead domains, companies that were acquired eighteen months ago, contacts who left, job titles that no longer exist, subsidiaries logged as separate targets, and accounts that fail the freshly validated ICP but survived from an older list. Every one of these dilutes the programme. Serving a tier-one experience to an account that should never have been on the list is not just wasted spend. It corrupts your reporting, because the denominator is wrong, and it erodes trust with sales the first time a rep opens a "priority" account and finds a shell.

Done properly, cleansing is more than deduplication. It is enrichment and structure: confirming the account still exists and still fits, mapping the real buying committee rather than a single stale contact, resolving parent and child company relationships so you are not marketing to the same organisation three times, and tiering the survivors by fit and by evidence of readiness. The list that comes out the other side is smaller, more accurate, and worth building a programme around. The one that goes in unexamined tends to cost you later, right when the results are due.

Only now do you run the programme

With those foundations in place, the parts everyone associates with ABM finally earn their keep. Tiering determines depth of investment, from one-to-one for the handful of accounts that justify it, through one-to-few for coherent clusters, to one-to-many for the broader validated set. Channels get chosen to match how each buying committee actually consumes information, whether that is programmatic, LinkedIn, content syndication, paid search, connected TV, or direct sales motion, usually in combination. Content gets built to speak to the specific problem your positioning claimed, at the stage each account is actually at.

This is the visible layer of ABM, and it works because of everything underneath it. Orchestration amplifies whatever it is pointed at. Point it at a confused message and a dirty list, and all you have done is scale the problem.

Measure against revenue, not activity

The last discipline is closing the loop back to the number that matters. ABM measured on impressions, clicks, and engagement scores will always look busy and often look successful, right up until someone asks about pipeline. The measures that count are account-level: penetration into target accounts, movement of named accounts into and through pipeline, influenced and sourced revenue, and velocity against your non-ABM baseline. This is where marketing and sales either share a definition of success or spend the quarter disagreeing about it. Agreeing the measurement model before launch is what keeps an ABM programme honest, and what lets you defend the budget when it is challenged.

The Spanb2b view

We built Spanb2b around a single idea: span the gap between marketing and revenue. ABM is one of the clearest places that gap shows up, and one of the most expensive places to get it wrong. Our starting position is that a large share of underperforming programmes never had a chance, because they began at the campaign and worked backwards, rather than beginning at positioning and the list and working forwards.

So we run the sequence deliberately. Positioning first, so the message is unmistakably yours. ICP validated against evidence, so the list is built on what actually converts rather than what looks good in a plan. The list itself cleansed, enriched, and tiered, so orchestration has something real to work with. Then, and only then, the programme, measured against revenue rather than activity.

None of this is the exciting part of ABM, but it is what makes the exciting part work. So if you are about to stand up a programme, or you are looking at one that spends well and returns little, resist the urge to blame the campaign. The problem is nearly always in positioning, the ICP, or the list. That is where we start, and it is usually where the fix is.