TL;DR
- The Problem: Marketing ops became a report factory. Teams measure the stack obsessively and still cannot route a signal to the one person who should act on it.
- The Cost: Most B2B content is never used by sales. The failure is not quality, it is routing: the right asset never reached the moment that needed it.
- The System: Capture, qualify, route, measure. Four layers that turn a pile of disconnected tools into a single pipeline.
- The Shift: AI made production cheap and routing expensive. Deciding what happens after a signal fires is now the scarce skill.
- The Start: Pick one signal, name one owner, set one SLA, and instrument the path end to end this week.
Every marketing organization I walk into has the same wall, and it is not a data wall. It is a routing wall. The team can tell you, to the decimal, how many impressions the blog earned last quarter, what the average email open rate was, and how many marketing qualified leads entered the funnel. Then you ask a simple question, “When a buyer signaled real intent last Tuesday, who saw it and what did they do?” and the room goes quiet.
That silence is the whole problem. Marketing operations was built to bring order to a growing stack: choose the tools, wire the data, own the dashboards, keep the lights on. That mandate quietly turned most ops teams into a report factory. They are excellent at describing what already happened. Almost nobody owns what should happen next. And in a market where AI has collapsed the cost of producing content and campaigns, the difference between a stack that compounds and a stack that just accumulates is not more data. It is routing.
The Report Factory Nobody Asked For
Walk through the average B2B marketing org and count the dashboards. There is one for paid, one for organic, one for email, one for social, one for the CRM, and a data studio board that pulls from all of them and is trusted by none of them. Each board answers a retrospective question. None of them answers the only question the business actually asks: what should we do, right now, about this specific account?
This is not a failure of effort. It is a failure of job description. For two decades the value of marketing ops was measured by how much could be quantified. The more you could report, the more sophisticated you looked. So the function optimized for legibility. Every new tool added a new reporting surface, and every new reporting surface created more work for the same finite team. The stack grew; the output did not.
The content numbers make the cost visible. Roughly 60 to 70 percent of B2B content never gets used by sales. The instinct is to blame the content: too long, too generic, too late. But talk to the reps and a different story appears. They did not reject the asset. They never found it at the moment they needed it. A case study that would have closed a deal sat three clicks deep in a content library while the rep rebuilt the same argument from scratch in a Google Doc. That is not a content problem. That is a routing problem wearing a content problem’s clothes.
A signal with no owner is not intelligence. It is a log line. The entire value of a signal is created or destroyed at the moment someone decides who acts on it and by when.
Why This Is a Routing Problem, Not a Data Problem
When teams discover that signals are dying in the stack, the reflex is to buy more data. More intent sources, more enrichment, more scoring. This is the same mistake as buying more shelves when the problem is that nothing gets delivered. You do not have a data shortage. Modern B2B teams are drowning in signal: content engagement, social interaction, product usage, hiring activity, funding events, competitor mentions. The scarce resource is not detection. It is disposition.
Detect is easy and getting easier. Disposition is hard, and getting harder, because disposition is an operations design problem. It requires someone to decide, in advance, which signals matter, who owns each one, what happens when it fires, and how fast. Most teams have never written those rules down. So the signal lands in a queue that nobody is watching, the moment passes, and the team reports on the volume of signals it captured as if that were the win.
This is also why the AI adoption gap hurts. Most marketers use AI ad hoc rather than as part of an integrated workflow, and nearly half of B2B teams have no usage guidelines at all. Ad hoc AI accelerates production and leaves routing untouched, which means it accelerates the pile-up of unused output. You get more assets, faster, for a delivery system that was already broken. The bottleneck moves from creation to disposition, and nobody has been hired to own disposition.
The good news is that a routing layer is not a new tool. It is a design decision you can make with the tools you already own, the same tools behind your first-party signal engine. The rest of this piece is the system I use to build it.
What Running Revenue Ops Taught Me About Signal
I have spent more of my career on the receiving end of the marketing to sales handoff than on the producing end, and it changed how I think about operations. When I was building social selling programs, the content that moved deals was almost never the content that won the internal awards. The winners were the assets that arrived in a rep’s inbox within an hour of a buyer doing something that mattered, with a one-line note explaining why now. Same content library. Completely different result. The variable was timing and ownership, not quality.
I have watched teams spend six figures on an intent data subscription and then route the output to a shared Slack channel that everyone muted by the second week. I have also watched a two-person ops team with a spreadsheet beat that subscription, because they wrote down four rules, assigned four owners, and made one person accountable for whether a signal was acted on within a day. Tools did not decide that outcome. The routing design did.
That is the lesson I would tattoo on the wall of every marketing ops standup: you are not paid to know what happened. You are paid to make sure the right thing happens next. Everything else is archival.
The Signal Routing System
Routing is not a slogan, it is a sequence. Four layers, in order, each one producing a decision. Build them once and the stack stops being a collection of tools and starts behaving like one system.
Most teams try to capture everything and end up acting on nothing. Start by naming the three to five signals that have actually preceded closed revenue in your data: a specific page visited twice, a product trial that hit a usage threshold, a champion who engaged three times in a week, a competitor mentioned in a call note. Write them down. Everything not on the list is reported, not routed. You can widen the net later, once the narrow one works.
Activity is noise that looks like progress. A blog view from a university IP is activity. A director of demand gen at a target account who read three posts, opened the pricing page, and matched your ICP is signal. Score on two axes only: ICP fit and intent strength. If a signal fails either axis, it goes to a nurture list, not a rep. Qualification is what keeps routing trustworthy, because a queue that gets spammed with false positives gets muted within a week.
Every signal gets one named owner and one deadline. Not a channel, not a team, an owner. Routed intent decays fast, which is why the clock matters more than the creative. Define three tiers: high-intent signals get a human touch within an hour, medium within a business day, low within a week. If a signal has no clear owner, the routing design is wrong, and no amount of volume will fix it.
The metric that matters is not how many signals you captured. It is the percentage of signals that were acted on within their SLA, and the percentage that produced a conversation. A routing rate of 90 percent on 100 signals beats a 10 percent rate on 10,000. Report this number next to revenue, because it is the closest thing marketing ops has to a leading indicator of pipeline.
The Four Layers, and How Each One Fails
Once the sequence is clear, it helps to name the layers so you can assign ownership and diagnose failure. Most broken ops stacks are not missing all four. They are strong in one and hollow in another, and the hollow layer quietly nullifies the rest.
| Layer | What it does | Failure mode | Metric |
|---|---|---|---|
| Capture | Detects the narrow set of signals that predict revenue | Captures everything, drowns the team, gets ignored | Signals captured per week |
| Qualify | Scores ICP fit and intent strength | Trusts volume and activity, floods reps with false positives | Precision (% of routed signals that were real) |
| Route | Assigns one owner and one SLA per signal | Routes to a channel, not a person, and dies in silence | Routing rate within SLA |
| Measure | Instruments the path from signal to conversation | Reports volume, never disposition | Signals to conversations |
Read the table top to bottom and the pattern is obvious. Teams rarely fail at capturing, because tooling makes capture easy. They fail at routing, because routing requires a human decision that no tool can make for them. That is where ops either becomes a system or stays a subscription list.
What Changes When Ops Owns Routing
When marketing ops stops reporting and starts routing, three things change at once, and each one is felt far outside the ops team.
First, sales stops searching and starts receiving. Instead of living in the CRM hoping to stumble onto a live account, reps get a short, curated list of accounts whose behavior says now. The content library stops being a graveyard and becomes a delivery mechanism, which is exactly how that 60 to 70 percent of unused content gets recovered. Second, content stops being judged on production and starts being judged on delivery: did this asset get routed to the moment it was built for? That single question changes the editorial brief overnight. Third, marketing stops looking like a cost center that produces artifacts and starts looking like an operating system that produces timing.
Reporting tells the business what happened. Routing changes what happens next. The second one is the job, and it is the one most ops teams were never asked to do.
The First 30 Days
You do not need a new platform or a reorg to start. You need one signal routed correctly, end to end, and then you replicate it. Here is the sequence I would run in the first month.
Choose the single signal that most reliably preceded a closed deal in the last year. Not the most interesting one, the most predictive one. Write its definition so precisely that two people would classify the same event the same way.
Assign a specific person to receive the signal and give them a deadline measured in hours, not days. Put both in writing. If you cannot name the owner, you have not finished designing the signal.
Log every occurrence from detection to the rep’s first action. You are measuring two things: routing rate and time to first touch. Do not add a second signal until this one runs without manual babysitting.
Compare routed signals to conversations. Kill it if it does not produce, or promote it and add the next signal. Routing is a compounding system, so the goal is never the first signal, it is the operating habit of deciding what happens next.
If you want a deeper view of the data side of this, start with why most marketing teams measure output instead of impact, then read why your attribution model is probably lying to you, and finally how signal decay should set your SLAs. The routing layer sits on top of all three.
Marketing ops was never supposed to be a report factory. It was supposed to be the nervous system of the revenue engine: sensing what matters and moving it to the place that can act. AI just made that mandate unavoidable, because when everyone can produce content, the teams that win are the ones that can deliver it to the right moment. Build the routing layer, and the stack you already own starts to look like a pipeline.
Stop asking your stack for another report. Ask it for a route. One signal, one owner, one clock, one metric. That is the whole game.
