If your pipeline is lumpy and your sales cycle is long, “Is this working?” is a trap question. Not because leadership is unfair. Because most marketing systems can’t answer it cleanly.
The awkward part: B2B marketing is full of ROI numbers that sound decisive—until someone asks how they were measured. One 2026 benchmark roundup claims SEO can drive 702% ROI with a ~7-month break-even (Oliver Munro). Another claims organic search generates 44.6% of all B2B revenue in its dataset (Oliver Munro). Email gets the same treatment: £36–£40 returned for every £1 spent (Oliver Munro).
Big numbers. Real sources. Still not a plan.
Because in the same 2026 data, 56% of B2B marketers say they struggle with attribution and 42% struggle with ROI tracking due to long sales cycles and complex attribution (Averi.ai). And a separate 2026 marketing-leader benchmark roundup says only 36% can accurately measure ROI, while 47% struggle with multi-channel ROI measurement (Genesys Growth). That’s the punchline: teams are asked for certainty in a system designed for ambiguity.
The one move: define “working” by motion, then instrument signals
Refine Labs’ core idea is simple and practical: don’t answer “is this working?” generically. Anchor it in the motion the initiative was built for—brand, demand, or expand—then show the best signal available now, plus what signal should come next if it keeps working.
It’s not semantics. It’s how you avoid debating the wrong scoreboard.
Motion-first framing also matches where B2B analytics is heading in 2026: closed-loop, revenue-focused attribution that ties marketing touchpoints and spend to CRM outcomes (pipeline and closed revenue), with deeper integrations across CRM, ad platforms, web analytics, and offline conversions (Directive Consulting; Salesforce positioning as summarized in the research brief). Tools are getting better. The question is whether the operating model is.
Here’s the tension to resolve: if leadership expects a revenue answer this week, and the channel’s break-even is measured in months, then the only honest path is a staged proof system. Leading indicators first. Lagging confirmation later. Same initiative, different clocks.
How to do it in practice (Step 1–3)
Step 1 — Name the motion (and write down the bet). Pick one primary motion per initiative. Not three. If it’s brand, the bet is preference and recognition with a specific audience. If it’s demand, the bet is qualified pipeline that converts. If it’s expand, the bet is post-sale growth signals (renewals, upsell, retention).
Short sentence. No hedging. This is where most teams quietly fail—everything becomes “pipeline” in a deck, and then everyone argues about why the pipeline didn’t show up yet.
Step 2 — Choose one “now” signal and one “next” signal. The “now” signal should move faster than revenue but still connect to the motion. The “next” signal is what you expect to see later if the “now” signal is real, not noise.
Examples that stay measurement-friendly without pretending last-click is truth:
- Brand motion: “Now” could be branded search lift or self-reported attribution trends (directional). “Next” is increased direct/organic-assisted pipeline in CRM over a longer window.
- Demand motion: “Now” could be cost per opportunity, win rate by source, or sales velocity by source. “Next” is closed-won revenue efficiency (CAC, LTV:CAC over time—once cohorts mature).
- Expand motion: “Now” could be upsell engagement or churn-risk signals (as defined by the business). “Next” is NRR movement after renewal cycles complete.
Step 3 — Forecast the next signal out loud. This is the underrated part of the Refine Labs framing: say what should happen next if the program is working, and when you expect to see it. Not fake certainty. Clarity.
And yes, put it in the report. When it’s written down, the conversation shifts from “prove it” to “did the expected signal show up?” That’s a much cleaner fight.
Run it this week: the “no-guess” readout template
Here’s the 5-minute version you can run this week:
- Owner: Demand Gen + RevOps (joint). Marketing writes the motion/signal logic; RevOps validates stages, pipeline definitions, and CRM joins.
- Tools: CRM + ad platforms + web analytics. If an attribution tool exists, use it for directional assist views, not as the judge and jury (Directive Consulting trend: closed-loop attribution tied to CRM outcomes).
- Timeline: 5 business days to implement; 2 weeks to get the first clean readout for “now” signals.
- Budget range: No new spend required to start. If testing is needed, cap at a low, predefined amount and treat it as measurement R&D.
Setup: Create a one-page table with columns: Initiative, Motion, “Now” signal, “Next” signal, Expected timing, Primary metric, Guardrails, Stop-loss.
The hypothesis (make it falsifiable): “If we report performance by motion and instrument one leading indicator plus one lagging outcome per initiative, then exec confidence will improve and reallocation decisions will get faster because the team will debate agreed signals instead of attribution opinions.”
Success = fewer ad hoc ‘prove it’ requests and faster budget decisions tied to the same definitions (qualitative but observable), plus improved ability to connect spend to CRM pipeline in closed-loop reporting over time (Directive Consulting trend). Guardrails = no redefining lifecycle stages mid-quarter; no claiming incrementality from platform dashboards alone. Stop-loss = if the “now” signal doesn’t move for two reporting cycles, pause spend increases and run a targeted diagnostic (creative fatigue, audience saturation, handoff breakdown).
What to measure (and what not to over-interpret): Use ROI benchmarks as context, not as targets. A 702% SEO ROI claim (Oliver Munro) doesn’t tell you your payback window, your content costs, or whether your CRM is clean enough to see the lift. Same with email ROI claims (Oliver Munro). Benchmarks can justify the category. Only your closed-loop data justifies the budget.
The trade-off (and when this is wrong)
The trade-off: this will reduce the amount of “easy” reporting. Some vanity wins disappear. Good. The point is to replace feel-good dashboards with decision-grade signals.
When this is wrong: if the business has a short sales cycle and high-volume transactional motion, you can often answer “is this working?” with tighter attribution windows and faster revenue feedback. In that world, motion-based reporting still helps, but the leading indicators matter less because lagging indicators arrive quickly.
For most mid-market and enterprise B2B, though, the 2026 measurement stats are the reality check: only 36% say they can accurately measure ROI (Genesys Growth). So the right move isn’t pretending to be in that 36% overnight. It’s building a system that earns confidence one signal at a time.
That’s the circle to close: “Is this working?” becomes answerable when everyone agrees what “this” was supposed to do, what “working” looks like this month, and what proof should show up next—before revenue finally arrives and takes all the credit.
