136 Days, and What ABM Is Actually For

Most marketing teams cannot tell you how long it takes to turn a first touch into a booked meeting. They can tell you spend. They can tell you leads, meetings, maybe pipeline. But the number that says whether the motion is actually working, and gives you something to improve, is the time from the first touch impression to the demo on the calendar.

For us right now, that is about 136 days. When the first touch is a paid impression, it is 107.

That single number reframes almost everything I do. It tells me paid is not a lead faucet we turn on and off by the week. It tells me the work I ship this quarter is mostly seeding meetings that land next quarter. And it gives ABM a job it can be held to instead of a vibe it gets judged on. Once you can name the cycle, you can go compress it, and in an ABM program you compress it with strategic, personalized outbound touches.

Why that number beats the ones on most dashboards

In B2B, paid can be a slow burn, especially against enterprise accounts. LinkedIn is our best paid channel, but it is primarily an influencing channel. Much of our demand capture happens on Google, paid and organic both. LinkedIn is where our ICP lives, and where we build influence over months. So the honest way to measure it is not "how many leads did the ad account produce last week." It is how long an account takes to move from the first impression we served to a real sales conversation.

Looking at prior-quarter closed-won in Dreamdata, 86% of those deals were influenced by paid channels before they ever landed in pipeline. The slow burn is doing real work. It just does it on a delay, which is why paid is so easy to underrate and then overcorrect on. A weekly lead count makes paid look like a bad month. A cycle-time view shows it doing exactly what it was built to do, one or two quarters upstream of the revenue.

Time from first touch to demo booked captures the thing every dashboard metric misses, which is momentum. Spend tells you what you put in. Leads tell you who raised a hand. Cycle time tells you how fast the whole system turns attention into intent into a meeting, and that is the only part you can actually pull on to grow faster.

Take one real month and look at it three ways on LinkedIn. The spend view says we invested a fixed budget and produced a handful of direct leads, which reads as expensive. The lead view says most of what the ad account touched has not raised a hand yet, which reads as underperformance. The cycle-time view says a batch of target accounts moved from cold to their first serious page visit, and a subset will surface as demos three or four months out. Same month, same spend, same accounts. Only the third view tells you what the money bought, and only the third view gives you a lever, because a timeline is something you can work to shorten while a weekly lead count is a number you react to.

The attribution trap

The delay is where teams get this wrong. In our Q1, 68% of closed-won came from pipeline created in earlier quarters. Sit with that. Most of the revenue we booked that quarter was seeded one or two quarters before it. The demand gen work driving this quarter's number mostly happened before the quarter started, and the work I am doing right now mostly seeds revenue that shows up in Q3 and Q4.

Measure demand gen by pipeline created and attributed inside the same quarter and you will undervalue everything that takes time to convert. Then you will starve the exact investment that feeds future quarters. It is a slow-motion own goal.

Picture the report that drives it. A quarter closes soft, same-quarter attribution says paid sourced very little, so budget moves to whatever produced a lead last week. The accounts paid had been warming for two months lose their air cover right before they would have converted. Ninety days later pipeline dries up, nobody connects it to the cut, and the cycle repeats. The teams that avoid this measure the lag directly, so a soft quarter reads as seeds still in the ground rather than paid failed.

This gets worse as you move up into mid-market and enterprise, where more decision-makers and longer evaluations stretch the lag further. And it runs this long for a well-understood reason. Only about 5% of B2B buyers are in-market in any given quarter, per the LinkedIn B2B Institute and Ehrenberg-Bass. Most of the impressions you serve today are planting memory for a purchase that could be quarters out. Les Binet's Share of Search work shows the same pattern from the other side, with brand demand taking six to twenty-four months to surface in the numbers.

What the dashboards cannot see

Even when the work converts, the tooling hides who did it. By the time a buyer acts the referrer is often gone, so dark social gets logged as direct. SparkToro put real numbers on how much traffic is misattributed this way. The channel doing the early work rarely gets the credit, and it looks weaker than it is on the report.

I hear the downstream version of this on almost every call with other marketers. Attribution becomes manual stitching across the CRM, ad platforms, Slack, and call recordings. The LinkedIn-ad-to-CRM handoff is misleading if you have no visibility before the conversion, so teams spending real money every month cannot see which contacts actually engaged the ad. Meanwhile the pressure to prove it comes top-down, from a new CMO, a CFO, or a board that wants a dollar figure. The most patient and most influential part of the funnel is also the least legible, which makes it the first thing questioned and the first thing cut.

None of that means paid is not working. It means the measurement most teams default to is built to miss it.

Cycle time sidesteps the whole fight. It does not ask which touch deserves credit, which is the argument that never ends and never changes a decision. It asks a simpler question. Across all of it, how long is the journey from first impression to booked meeting, and is that number getting shorter. You do not need perfect multi-touch attribution to answer that. You need a start point you can define, an end point you can count, and the discipline to measure the same window every quarter. That is a bar most teams can clear, which is why it beats the more sophisticated report nobody trusts.

So what is ABM actually for

Once you accept the slow burn, the job of ABM comes into focus. You are deliberately moving accounts along a progression, from identified to aware to interested to considering and finally to closed-won revenue. You are trying to shorten the gap between the impression that created interest and the meeting that acts on it. Rather than wait 136 days for an account to wander to a demo on its own, you watch for the behavior that says an account is paying attention, and you reach out inside that window. Detect the signal, reach your buyers, book the meeting.

The signals are specific, not vague. A cluster of visits to a high-intent page. Someone researching a competitor. A job change into a buying role. Repeat ad engagement from named contacts at a target account. These are the moments an account tips from being marketed to toward being worth a direct conversation, and the whole point of ABM is to catch them while they are live instead of finding out later in a pipeline report.

This is where site de-anonymization does real work for us. We de-anonymize the contacts landing on our high-intent pages, then act on that behavior with a workflow that is almost embarrassingly simple. A few filters on a segment, notifications into Slack and the CRM, and a webhook. When a named contact from a target account shows the behavior we care about, the webhook sends the lead to get enriched and routes it straight to the person who should reach out, while the interest is still warm. Each AE has their own approach for opening the conversation.

That is the whole trick. Act on a signal you created, before a buyer is ready to fill out a form but while they are interested enough to warrant a conversation. Done consistently, that is how you increase the velocity of closed-won deals.

The economics are the part that surprises people. Reaching one named buyer at the exact moment they are paying attention used to be a 1:1 effort you could only justify for a handful of logos. Now that same precision runs across a whole target list at 1:few cost, because the software does the watching and the routing. The outbound itself is better because it is specific. Not "just checking in," but a message that maps to what the buyer was actually researching, which is the difference between a reply and a delete.

The hardest part is translating the signal into something meaningful. There is a real difference between saying "saw you were on our website" and tying a buyer's research back to the pain they are trying to solve.

How we are pressure-testing 136 days

A number you cannot move is a vanity metric with extra steps. So we built a way to test it.

We started serving ads to our target account list in the first week of June. One cohort gets 1:1 personalized campaigns and will receive outbound touches after a threshold of engagement. Another cohort gets more generic content and our regular treatment, which means we engage them at sign-up or demo. Run the 136-day average forward and the earliest of these accounts start turning into demos around December 4th. That is the baseline. The accounts left to convert on their own, on the natural slow-burn timeline, are the control.

Against that, we run the signal-based outbound motion on accounts that trip a signal. De-anonymize, detect the behavior, reach out inside the window. The comparison we care about is the gap between the two groups. If the motion works, the accounts we act on book demos meaningfully faster than the 136-day baseline the slow burn would have produced on its own, and we can read the difference in days rather than vibes. We will also watch which signals actually precede a booked meeting, so the filters sharpen over time instead of firing on noise.

The experiment measures one thing. How much we can compress the time from first impression to booked meeting, and by how many days. Leads produced last week was never the right question to ask of a slow-burn channel. This is a claim we can hold ourselves to, which is the only kind worth making.

136 days is already a strong number

For context, 136 days from first impression to booked demo is good, not bad. The more commonly tracked figure, the post-opportunity sales cycle, sits around an 84-day median and has lengthened roughly 22% since 2022, with every added decision-maker stretching it further.

Those two numbers measure different windows, which is the point. Most teams only measure the late stage, after an opportunity already exists, and never put a number on the long stretch before it. Plenty of teams sit closer to a year end to end, if they can measure it at all.

We can name ours across the full window, from the first ad we served to the meeting we booked. That is the only reason we can go compress it.

If you take one thing from this

Measure the cycle. You cannot compress a timeline you have never put a number on, and you cannot defend the spend that feeds it if you only judge it one quarter at a time.

Put a number on the full window, from first impression to booked demo, and two things change at once. Paid stops looking like a bad week and starts looking like the slow-burn engine that accelerates your booked-meeting velocity. And ABM gets a job it can be held to, which is to find the accounts already warming up and reach them before the slow burn would have gotten there on its own.

Ours is 136 days. Now we go make it shorter, one de-anonymized signal at a time.