Every chapter in Part Two included a measurement section (specific metrics for specific functions). This chapter is about measuring the system as a whole, not just the parts.
Because here’s the thing about connected systems: the sum is greater than the parts.
A blog post’s goes beyond pageviews. It’s the outbound email that attached it, the ABM landing page that featured it, the prospect who read it before a sales call and walked in already trusting you.
A sales call’s value isn’t just the deal it progresses. It’s also the insight that fed a blog post, the customer language that improved outbound messaging, and the signal that triggered an ABM escalation.
Traditional marketing measurement was designed for disconnected functions. Content tracks pageviews, demand gen tracks MQLs, sales tracks pipeline… each function optimizing its own metrics with nobody measuring the connections between them.
When you build a systems-led growth engine, you need a different measurement framework that captures the compounding effects, the cross-functional value, and the system-level health that individual metrics miss.
§What to Stop Optimizing For
I’m not saying to stop tracking any one of these metrics entirely. Some of them are useful as diagnostic signals. But they should not be the metrics you optimize against, because optimizing for them will pull you away from the system’s actual purpose: building pipeline.
1. Pageviews: I’ve made this case already, but it bears repeating because pageviews are the most seductive metric in marketing. They go up and look great on dashboards. They’re easy to explain to executives who aren’t experts in what metrics should matter to marketers. And they tell you almost nothing about whether the right people are finding you or whether they’re doing anything about it. I cut traffic from 350k to 210k and pipeline went from zero to literal millions. If I’d been optimizing for pageviews, I would have optimized for failure.
Track pageviews as a diagnostic (is traffic trending up or down by segment?), but:
- Don’t optimize for them
- Don’t set quarterly goals around them
- And don’t celebrate them in team meetings.
2. Impressions and follower count: The same logic applies. A LinkedIn post that gets 50,000 impressions from the wrong audience is less valuable than one that gets 2,000 impressions from decision-makers at target accounts. A CEO with 50,000 followers who are mostly other marketers has a less useful audience than one with 3,000 followers who are all potential buyers. Impressions measure exposure, but they don’t measure impact.
3. MQLs as traditionally defined: If your MQL definition is “downloaded a PDF” or “visited three pages in a session,” you’re sending sales leads that are curious, not leads that are buying. The traditional MQL was designed for a world where marketing’s job was to fill the top of the funnel and sales’ job was to qualify. In a systems-led model, the scoring happens based on behavioral signals and enrichment data, not on arbitrary engagement thresholds. Replace MQLs with the intent-scored leads from Chapter 7.
4. Content volume: Publishing 30 blog posts per month is not inherently better than publishing 15. Publishing 15 ICP-focused, AEO-optimized, human-reviewed articles that are tagged in the content library and connected to outbound and ABM workflows is dramatically better than publishing 30 generic posts that aren’t connected to anything. Volume without system integration is noise (on the other hand, if you can do 30 ICP-focused, AEO-optimized, human-reviewed articles, then have it… this is one of the only areas in marketing where more can actually be better if quality doesn’t suffer for it).
5. Activity metrics without context: “We sent 500 outbound emails this week” tells you nothing. “We sent 50 personalized emails to enriched target accounts and booked 6 meetings” tells you everything. Activity without context is the refuge of teams that are busy but not productive.
§What to Start Tracking
These are the metrics that capture how the system is actually performing. Some are familiar and some are new, but all of them measure outcomes or system health rather than activity.
1. Pipeline velocity
How fast are deals moving through your pipeline? This is the metric that most directly reflects the health of the whole system. When content is reaching the right people, when outbound is relevant, when inbound responses are fast and contextual, when thought leadership is building trust, when ABM is targeting the right accounts, and when case studies are reducing buying risk, pipeline moves faster. If velocity is declining, something in the system is broken. The diagnostic then becomes: which function’s metrics are lagging?
Pipeline velocity is calculated as:
(number of qualified opportunities x average deal value x win rate) divided by average sales cycle length.
Track it monthly and trend it over quarters. A healthy system shows velocity improving over time as the compounding effects accumulate.
2. Content utilization rate
Of all the assets in your structured content library, what percentage was used by at least one workflow, one sales rep, or one prospect in the last 90 days? This is the single best measure of whether your content strategy is aligned with your go-to-market needs.
- Below 40%: your library has a findability or relevance problem.
- 40% to 60%: working but with gaps.
- Above 60%: strong alignment.
- Above 80%: exceptional.
Track which assets get used most and which categories have gaps (no healthcare case studies, no competitive positioning for your biggest competitor, no content addressing the implementation concern that comes up in every sales call). Those gaps are your content production priority list.
3. System efficiency (the input-to-output ratio)
How many publishable assets does one input produce?
- A podcast episode should produce 10 or more.
- A customer interview should produce 6 to 8.
- A sales call should produce at least 3 (follow-up email, tagged insights, account profile update).
Track this ratio across input types. If it’s declining, the repurposing workflows need attention. If it’s improving, the system is getting smarter (better templates, richer library, more refined prompts).
This is the metric most unique to systems-led growth. Traditional marketing doesn’t track it because traditional marketing doesn’t produce multiple outputs from a single input. When you start tracking it, you see the efficiency advantage of the system in concrete terms.
4. AEO visibility
How often does your brand appear in AI-generated answers for queries relevant to your market? This is the emerging metric that most companies aren’t tracking yet, which means tracking it gives you an information advantage. Use whatever tools are available (HubSpot’s AEO Grader, manual monitoring, specialized platforms). If it’s growing, your content and AEO strategy are working. If it’s flat while competitors are growing, you have a gap.
5. Deal influence by content type
For every closed-won deal, look back at the content engagement history.
What did the buying committee consume during their evaluation?
- Did they read blog posts?
- Engage with thought leadership?
- View case studies?
- Visit an ABM landing page?
- Reference content in a sales call?
This data tells you which content types have the most influence on actual deals, not just which ones generate the most traffic. I’ve consistently found that case studies and competitive positioning content have outsized influence on closed deals relative to their traffic volume. Blog posts drive awareness, whereas case studies drive decisions.
6. Customer language adoption rate
This one is harder to measure but profoundly important. Are the phrases and descriptions from your customer language database showing up in your outbound messaging, your blog content, your ad copy, and your sales scripts? When a prospect says, “Your email described exactly what we’re dealing with,” that’s customer language adoption working. Track it qualitatively (ask reps if prospects resonate with the messaging) and directionally (A/B test outbound sequences using customer language versus marketing language and measure response rates).
§The Compounding Metric
There’s one question that captures the health of the entire system better than any individual metric:
Does each new input make the system better? If yes, the system is compounding.
- Every new blog post enriches the content library.
- Every sales call adds to the customer language database.
- Every case study expands the proof portfolio.
- Every event produces derivatives that reach new people and generate new signals.
The system gets smarter, more contextual, and more effective with every cycle.
When the answer is no, something is disconnected.
- Maybe the content engine is producing articles but they’re not getting tagged and stored.
- Maybe sales calls are generating insights but those insights aren’t flowing back to the content team.
- Maybe case studies are being published but they’re not being surfaced by the outbound and ABM workflows.
The system is running, but it’s not compounding.
The diagnostic is straightforward. Trace a recent input through the system and see how far it travels. Take the last podcast episode you recorded:
- Did the transcript get processed?
- How many assets were produced?
- Were they tagged and stored in the library?
- Did any of them get used by a sales rep?
- Did any of them appear in an outbound sequence?
- Did any of them show up on an ABM landing page?
If the input traveled through three or more workflows, the system is connected. If it stopped after one, you have a plumbing problem.
I run this trace once a month. I pick one recent input (a blog post, a sales call, a podcast episode, a case study) and follow it through the system. Where did it go? What did it produce? How many other workflows touched it? This simple exercise reveals more about system health than any dashboard.
§How to Report This to People Who Care About Revenue
Most executives don’t care about content utilization rates or AEO visibility scores. They care about pipeline, revenue, and efficiency. Here’s how to translate the systems metrics into language that resonates in a board room or a quarterly review.
1. Pipeline
“Our content-influenced pipeline grew from $X to $Y this quarter. Here are the three content types that had the most influence on closed deals.”
Lead with the number. Follow with the specifics that explain it, but don’t show them the content calendar when you can show them the pipeline.
2. Efficiency
“One person is producing the pipeline-driving content output that previously required a team of four, at a total infrastructure cost of $X per month.”
This is the skeleton crew argument in metric form. Executives understand headcount economics, so show them the comparison between what the system produces and what it would cost to produce the same output with humans.
3. Velocity
“Our average sales cycle shortened from X days to Y days this quarter. We attribute this to faster inbound response times, more relevant outbound personalization, and prospects arriving at sales calls better educated from our content.”
Velocity connects the system to revenue timeline, and faster velocity means faster revenue recognition. CFOs care about this deeply.
4. Compounding
“Each piece of content we produce now generates X assets across Y channels, compared to Z assets six months ago. Our cost per asset has decreased by Q% while our pipeline has increased by R%.”
This is the systems argument: the infrastructure we built is producing increasing returns over time. It’s the opposite of the linear model where output scales with headcount.
Keep the system-level metrics for your own operational review. Translate them into pipeline, efficiency, velocity, and compounding for the people who sign off on budgets.
The system is the how; pipeline is the what. Report the what and use the how to explain it.
Attribution is still hard. I want to be direct about this because it’s the place where the measurement story gets messiest.
In a connected system, a closed deal might have been influenced by:
- A blog post the prospect read three months ago.
- An outbound email that referenced a case study.
- An ABM landing page that featured industry-specific proof.
- A meeting prep brief that made the rep more effective.
- A follow-up email that was generated from the call transcript.
Which of those “caused” the deal? All of them together, and none of them individually.
Multi-touch attribution models try to solve this by distributing credit across touchpoints. First-touch, last-touch, linear, time-decay, position-based. Each model tells a different story, but they’re all approximations.
I’ve settled on a pragmatic approach: influence attribution rather than cause attribution. I don’t try to determine which touchpoint “caused” the deal. I track which content the buying committee engaged with during their evaluation period and report it as “content that influenced the deal.” This gives me directional data about which content types and topics matter most, without pretending I can assign precise credit to individual assets.
For AEO specifically, the attribution gap is even wider. When a prospect asks ChatGPT about your product category, gets a recommendation that includes your brand, and then visits your website directly, the CRM shows a direct visit. There’s no referral tag from ChatGPT (yet), and people’s personal interactions with each LLM can impact responses in ways no one totally understands yet. AEO influence is invisible to traditional analytics. You know it’s happening because your AEO visibility is growing and your direct traffic is increasing in correlation, but you can’t (totally) prove the causal link with standard tools.
This is uncomfortable for someone who preaches pipeline over pageviews. I’m asking you to invest in a strategy (AEO) where the attribution is fuzzy and the ROI is directional rather than precise. I’m doing it because the trajectory of buyer behavior is clear (more AI-mediated discovery, fewer traditional search clicks), and waiting for perfect attribution before investing means arriving after the window has closed.
Measure what you can. Be honest about what you can’t. And make decisions based on directional conviction when precise measurement isn’t yet possible.
That speaks less to the failure of the measurement framework and more to the reality of operating at the frontier of new tech.
§The Dashboard That Matters
If I could only look at five numbers each month to assess the health of the system, they’d be these:
1. Pipeline velocity (is it improving?)
2. Content utilization rate (is the library being used?)
3. Input-to-output ratio (is the system efficient?)
4. AEO visibility (are we showing up in the new discovery layer?)
5. The compounding trace (did the last input travel through the system?)
These five tell me whether the factory is running, whether the pipes are connected, and whether the chocolate is flowing.
The next chapter is the last one: what happens when the pipes are in place, where this is all going, and why there is no shortcut for the work that only you can do.