Account-Level Signals Aren’t Worthless. They’re Early.

A CRO posted on LinkedIn last week that account-based marketing is an excuse to do lazy marketing, and that account-level signals are absolutely worthless. I read it twice. I’m a salesperson. I’ve built these programs for myself, in long-cycle B2B, against named buying committees, with real budget on the line. Calling that work lazy is the lazy take.

The frustration underneath the post is fair. Most ABM programs are weak in their first cycle. Most dashboards inflate the sense of activity. Marketers do hide behind vanity metrics, and salespeople do waste time chasing noise. But the conclusion (the whole approach is worthless) usually comes from two specific confusions: misreading leading indicators as if they were lagging ones, and conflating account-based reporting with actual ABM. Once you separate those, the program either starts working, or you find out faster that it shouldn’t be running at this account in the first place.

This is how I think about it from a seller’s chair.

TL;DR

• Roughly 80% of a long B2B buying cycle is anonymous. Account-level signals are not lazy. They are the only signal available during that window.
• Form fills, replies, and meeting requests are lagging indicators. By the time they fire, the buyer has been watching you for 100 to 150 days.
• Account-based reporting (broad list, company-level signals) and true ABM (10% of the list, named humans, deliberate plays) are two different programs. Most “ABM is dead” takes are aimed at reporting.
• Even the best stack (HubSpot, ZoomInfo, RB2B, Vector, Clay) resolves only 30 to 40% of buyers to a named individual in the anonymous window. That is a strong result, not a failure mode.
• The goal of ABM is to qualify or disqualify accounts faster than your average cycle. Not lead volume. Not dashboard activity.
• If sales does not handpick the 25 accounts, validate the buying committee per account, and stay in the program across a full cycle, the laziness is sitting on the sales side.

1. Most “ABM is lazy” takes are really lagging-indicator takes

The complaint that account-level signals are worthless usually means the complainant is grading them as if they were ready-to-buy signals. They aren’t. They’re leading indicators in an environment where the buyer is structurally invisible for 100 to 150 days, and they’re the only signal you’re going to get during that window.

Why it matters

In a long sales cycle, roughly 80% of the journey happens before the buyer is willing to let you see them. Email security gateways break click tracking. Bots inflate visit counts. Corporate VPNs hide IPs. Cookie opt-out is legally required in most markets and widely enabled at the corporate machine level. The buyer is moving through research, internal conversations, and vendor comparisons, and none of it shows up as a form fill. When your website visitor tool says Acme Corp had 13 visits across 28 pages for 35 minutes in the last 30 days, that’s not a closing argument. It’s a leading indicator that someone over there is paying attention, and your job is to figure out who. Dismissing the signal because it didn’t arrive pre-qualified is dismissing the only data the buyer is willing to give you in that window.

How to use it operationally

• Treat each account-level surge as a trigger for de-anonymization work, not as a closing event. The signal is step one in a process, not the finish line.
• Run the de-anonymization stack honestly. HubSpot and ZoomInfo resolve traffic to the company, not the individual. RB2B and Vector can reach individual visitors but only when they resolve, which is inconsistent. Clay aggregates and enriches across providers. Across all of them, getting to 30 to 40% individual identification in the anonymous window is a strong result, not a failure.
• Correlate the signal back to your active programs. If visits spike the week after an email send, someone on that list drove it. If they correlate with a LinkedIn matched-audience campaign, you know which named cohort to lean on.
• Reserve lagging indicators (form fills, replies, meeting requests, deal creation) for what they are: confirmations from buyers who have already been watching for months. They tell you the process is real. They do not tell you when the process started.

Watch-outs

• A single email open or first content download is a leading indicator. Repeated engagement with consideration-stage assets (vendor comparisons, ROI calculators, integration guides) starts to behave more like a lagging indicator. Calibrate the response to the asset, not the action.
• Most B2B programs should ungate their content. Unless you’re Gartner, Forrester, or The Economist, gating throttles the same pipeline you’re complaining is empty.
• “Account-level signals are worthless” is almost always an attribution problem in disguise. The signal isn’t worthless. The system around it isn’t doing the de-anonymization work the signal exists to trigger.

2. What you’re calling ABM probably isn’t ABM

Almost every “ABM is dead” post is a complaint about account-based reporting wearing an ABM label. True ABM is a much smaller and much more expensive program, and most teams aren’t running it.

Why it matters

Account-based reporting is the broad layer. It runs across your full target list, often 2,500 to 5,000 accounts, and tracks company-level signals: website visits, ad engagement, email opens aggregated by account. By definition it’s noisy and largely anonymous. Its job is to flag which accounts deserve consideration for true ABM next cycle. True ABM is the focused layer. It runs against roughly 10% of the list, with the buying committee named, using deliberate tactics: one-to-one display (Influ2-style person-level advertising), LinkedIn matched audiences scoped so only those 8 named people at Acme Corp see the campaign, executive gifting, direct mail, custom landing pages, deliberate plays at conferences where their executives are speaking. When a CRO calls ABM lazy, they’re almost always grading the reporting tier with ABM’s job description.

How to use it operationally

• Size the program honestly. Five reps, 25 accounts per rep per cycle, gives you a 125-account ABM cohort. The one-to-many channels (email blasts, newsletters, broad ads) still go to the larger 2,500 to 5,000 universe, because reps will inevitably eliminate accounts mid-cycle (the seller plays golf with Acme Corp and finds out they’re locked into a competitor through 2028) and need backfill.
• For every ABM account, name the buying committee and verify the contacts before any outreach. Eight people per account times 25 accounts is roughly 200 humans. That is a reasonable research load for a seller who actually owns those accounts.
• Use channels that resolve to named people. Influ2 for one-to-one display. LinkedIn matched audiences locked to the named contacts only, so you know any engagement came from those specific 8. Individual email campaigns where those 8 are the only recipients. Per-account landing pages that exist for them and no one else.
• Keep account-based reporting and ABM separate in dashboards, meetings, and expectations. They feed each other, but they answer different questions.

Watch-outs

• If you can’t list the named contacts in an ABM account, the program is reporting wearing an ABM label.
• “We do ABM” against 2,500 accounts is not ABM. Even the most sophisticated programs in the market don’t run true ABM at that scale.
• Calling it “account-based selling” instead of “account-based marketing” can help when sales acts like ABM is a marketing-only problem. The label matters less than the shared ownership model.

3. The seller does half the program

ABM is a feedback loop between sales and marketing across at least one full sales cycle, often two. If sales sits out the program, the program looks exactly as lazy as the CRO in the LinkedIn post says it is. That’s the part nobody wants to print.

Why it matters

Garbage in, garbage out applies to ABM more than to any other channel, because the program runs against a small enough list that one weak input poisons the quarter. If you hand marketing a list of unvetted accounts and then complain that the signals are noisy, the program is failing on the inputs, not on the approach. The first one or two cycles are calibration. A 200-day sales cycle means 200 to 400 days before you can credibly say which signals actually correlate with opportunity creation in your environment. Killing the program at 60 days guarantees it never gets to be anything else. I’ve built these programs for myself. Sitting on the sidelines and calling them lazy is, frankly, abdicating the role.

How to use it operationally

• Spend a real week or two researching the 25 accounts before any outreach. Provide marketing the curated buying committee and the reasoning behind it. This is the seller’s job, not marketing’s.
• Lock the active ABM list for one full sales cycle. Allow a couple of validated replacements per quarter, no more. Rotation is for real disqualification (a confirmed competitor contract, a confirmed not-in-market signal), not for impatience.
• Run a sunset rule on the reporting layer. Accounts showing zero engagement across a full cycle move to the back of the line so budget compounds against accounts that are actually warming. Zero engagement over a full cycle is information.
• Define the program goal as one explicit measure: are accounts qualifying or disqualifying faster than the average cycle? That’s what ABM is accountable to. Not opens. Not site visits in isolation. Speed to a decision.

Watch-outs

• A seller who wants to replace their full ABM book at day 60 is reporting a seller-discipline problem, not a program problem.
• Mistaking a marketing-qualified account (MQA) for a marketing-qualified lead (MQL). An MQA is a company showing aggregate engagement. An MQL is a named human who has done something attributable. They require different responses and shouldn’t be reported as the same thing.
• Starting ABM before sales and marketing have working trust. If the relationship is adversarial, the feedback loop never closes and the program never calibrates. Sort that out first.

Context on Outkeep’s Approach

Outkeep operates in long-cycle, high-consideration B2B, where most of the buying journey happens before a buyer is willing to be seen. We spend time on the distinction between account-based reporting and ABM because the failure mode is the same one we see in deliverability work: teams declaring something dead because they’re measuring it with the wrong instrument. A de-anonymization workflow built on an email program with poor deliverability or unreliable tracking is built on sand. The strength of leading indicators depends on whether the underlying infrastructure is trustworthy, and getting that foundation right is what makes account-level signals worth acting on.

Across the industry, what’s actually broken is the patience and the participation, not the program design. The teams running ABM well are the ones whose sellers research their own accounts, name their own buying committees, and commit to a full sales cycle before grading the program. That’s an operating model, not a tool stack.

FAQ

Why do most B2B buyers stay anonymous for so long?

Email security gateways break click tracking. Bots distort website analytics. Corporate VPNs hide IPs. Cookie opt-out is legally required and widely enabled at the corporate machine level. Buyers also complete most of their research before identifying themselves, because identifying yourself triggers a sales process they aren’t ready for. The result is that 80% of a long B2B cycle happens anonymously, and even the best de-anonymization stack resolves only 30 to 40% of visitors to a named individual during that window.

What’s the difference between account-based reporting and account-based marketing?

Account-based reporting is the broad layer (2,500 to 5,000 accounts, company-level signals, leading indicators). Its job is to flag which accounts deserve consideration for ABM next cycle. Account-based marketing is the focused layer (typically 10% or less of the list, named humans, deliberate plays). It runs against people, not company logos. Most “ABM is dead” critiques are aimed at reporting.

What’s the difference between an MQA and an MQL?

A marketing-qualified account (MQA) is a company showing aggregate engagement across channels. A marketing-qualified lead (MQL) is a named individual who has done something attributable. They are not the same signal and should not be reported as the same thing. An MQA without an attached named lead is a prompt to do de-anonymization work, not a closing argument.

What tools actually de-anonymize buyers in the anonymous window?

HubSpot and ZoomInfo resolve web traffic to the company level. RB2B and Vector can reach individuals when they resolve, though resolution rates vary. Clay aggregates and enriches across multiple providers. Influ2 is the standard tool for one-to-one display against named contacts. None of the resolution tools get above roughly 30 to 40% individual de-anonymization during the anonymous window, and that’s a strong result.

How long should you commit to an ABM program before evaluating it?

At least one full sales cycle, ideally two. A 200-day cycle means 200 to 400 days before you have enough data to tell which signals actually predict qualification in your environment. The first cycle is calibration. You will get false positives and false negatives. Rotating accounts every 60 days means the program never generates usable data.

What is the actual goal of ABM?

To qualify or disqualify accounts faster than your typical cycle average, and to shorten the anonymous portion of the buying journey. ABM is not designed to generate a high volume of leads. It’s designed to move a small, prioritized set of accounts to a decision faster than they would reach one on their own.