Cracking the Sales Management Code by Jason Jordan and Michelle Vazzana: Summary, Key Lessons, and RevOps Takeaways

Two ways to run a sales organization: results-first versus activity-first

Almost every sales organization is running one of two operating models, whether or not anyone has named it. Understanding which one you have is the precondition for using anything in this book.

The results-first model is the default. Leadership sets a revenue number, cascades it down as quotas, builds a dashboard whose headline tiles are bookings, ARR, pipeline value, and attainment-to-plan, and then holds managers accountable for those tiles at a weekly forecast call. The manager's job, under this model, is to *explain variance*. When the number is short, the meeting becomes a search for reasons — a deal slipped, a competitor discounted, procurement stalled. The manager leaves with an obligation ("get it back next month") and no specified mechanism. Coaching, when it happens, is deal-by-deal firefighting on whatever opportunity is largest and closest to the line.

The activity-first model inverts the reporting structure without abandoning the number. Revenue still exists on the dashboard, but it is treated as a readout, not a control. The tiles a manager is actually held to are things like qualified conversations held, opportunities advanced past a specific stage, account plans refreshed, target accounts touched this quarter. The forecast call still happens, but the *management* call is separate and is about inputs. When the number is short, the diagnostic question isn't "why?" but "which input fell, and when did it fall?"

Jordan and Vazzana's contribution is not the vague advice to "focus on leading indicators" — that idea predates them. It's the insistence that leading indicators come in two distinct grades that behave completely differently, and that conflating them is why most leading-indicator programs quietly fail. Sales objectives — customer acquisition rate, retention rate, share of wallet, average deal size, sales cycle length — feel manageable because they're upstream of revenue, but they are not. A rep cannot decide to retain a customer. They can decide to make the call. Objectives are *influenced*, through where you point the sales force strategically. Sales activities — calls, meetings, proposals, demos, account plans, pipeline hygiene — are the only tier where a manager can pick up the phone today and change the number by asking.

The authors offer a clean test for the boundary: if you cannot ask a rep to do more of it tomorrow, it is not an activity. "Send five more proposals" passes. "Close two more deals" fails — the rep does not control the buyer's signature. "Improve retention" fails. "Complete a QBR with your top ten accounts this month" passes, and it is the retention lever expressed as something a human can actually execute.

The trade-off between the models is real and worth naming honestly. Results-first is cheap to run — the data already exists in your CRM as closed-won records, and it requires almost no instrumentation. Activity-first requires you to capture, clean, and trust activity data, which is expensive and historically has been the failure point: reps under-log, managers stop trusting the dashboard, and within two quarters everyone drifts back to staring at bookings. The book was written before automatic activity capture made this cheap; that shift is the single biggest thing that has changed in its favor.

Choosing between the models — and which processes to instrument first

The decision isn't binary in practice. Nearly every organization runs a hybrid: results for the board, objectives for the annual plan, activities for the weekly one. The real decision is *where you spend your instrumentation budget and your managers' hours*, and that depends on your motion.

Start by identifying which of the four core processes actually carries your revenue. Jordan and Vazzana map them as call management (the individual interaction), opportunity management (the deal moving through pipeline), account management (growing what you already have), and territory management (covering a market). The mistake is trying to instrument all four at once. A high-velocity inside-sales motion lives or dies on call management and early-stage opportunity management; territory planning is nearly irrelevant when leads route automatically. A field enterprise motion with eighteen-month cycles and forty named accounts per rep is almost entirely account and territory management; raw call volume tells you nothing useful. A land-and-expand SaaS org needs opportunity management for the land and account management for the expand, and will get burned if it uses the same dashboard for both.

Here's a decision flow for picking the primary process and the metric grade you manage to:

The sequencing principle underneath that flow is restraint. Instrument two or three activities per process, not fifteen. The failure mode of an activity-first rollout is a dashboard with forty input tiles, which is functionally identical to the results-first dashboard it replaced — nobody knows which lever to pull, so they pull none. Pick the two activities you would coach in a one-on-one if you only had ten minutes, and put those on the wall.

The second sequencing principle: build the cascade, not the tile. A retention number that drops is useless on its own. A retention number that sits next to "QBRs completed" and "executive touches logged" for the same segment, over the same trailing period, tells a manager whether the problem is effort or positioning. That activity-to-objective-to-result chain is the actual deliverable of a RevOps team applying this book — not a longer metric list, but a shorter one arranged causally.

What the numbers look like in each model

Specifics matter here more than principles, because the models diverge most visibly in how time and data get allocated.

Manager time allocation. The book's argument is that effective frontline managers spend the clear majority of their time on activity management — coaching, ride-alongs, call reviews, pipeline inspection focused on what the rep does next — rather than on reviewing results reports or personally rescuing deals. In a results-first org, the inverse is typical: the manager's calendar is dominated by forecast prep, roll-up meetings, and deal escalations. A useful audit is to categorize a manager's last two weeks of calendar into result-review, deal-rescue, and activity-coaching buckets and look at the split. Most teams discover activity coaching is the smallest slice, often squeezed into whatever survives after forecast season.

Span of control. A frontline sales manager typically carries somewhere in the range of six to twelve direct reports. That number is the hard constraint on the whole model: if each rep is working dozens of accounts and multiple live opportunities, the manager is nominally accountable for hundreds of revenue streams they cannot personally touch. Activity management is what makes that span survivable — you cannot inspect three hundred deals, but you can inspect whether ten reps each held their target number of qualified conversations and whether the ones who missed share a pattern.

Metric counts. A typical enterprise CRM instance exposes dozens to hundreds of reportable fields, and mature orgs routinely build dashboards with thirty-plus tiles. The book's prescription collapses that to a working set of roughly five to nine managed metrics per role — a couple of activities, one or two objectives, and the results they roll into. The discarded metrics aren't deleted; they move from the management dashboard to the analysis layer, where RevOps uses them for diagnosis rather than accountability.

Stage conversion versus win rate. In opportunity management, the model swap has a concrete data expression. Results-first reports overall win rate, a single number that arrives too late to act on. Activity-first reports stage-to-stage conversion, which localizes the leak. If discovery-to-demo converts well but demo-to-proposal collapses, you have a demo problem, and that's coachable this week. Pair it with stage aging: flag any opportunity sitting in a stage longer than the median dwell time for that stage in that segment, and route it to a manager review queue. This is a genuinely mechanical RevOps build — a rolling median, a comparison, an alert — and it converts a lagging win-rate conversation into a weekly working list.

Activity thresholds. The authors deliberately avoid publishing universal activity targets, and that restraint is correct. The right number of qualified conversations per week for a transactional SMB rep and for an enterprise field rep differ by an order of magnitude. What transfers is the method: measure your own top-quartile performers' activity levels over a trailing period, set the floor there, and re-derive it every couple of quarters as the motion changes. Borrowed benchmarks from another company's motion are worse than no benchmark, because they carry false authority.

Qualification as a filter, not a script. On call management, the meaningful count is qualified conversations, not dialed calls. Whatever qualification framework you use — budget, authority, need, timeline, or a modern variant built around problem confirmation and access to power — the point is that the activity metric embeds a quality bar. Counting raw dials produces exactly the behavior you'd predict: more dials, worse dials. Counting conversations that cleared a defined bar makes the metric harder to game and worth coaching to.

Building it: instrumentation, cadence, and the account and territory layers

The implementation problem is that activity data is the most expensive data in the stack to make trustworthy, and the cadence built on top of it is the most fragile habit in the management system. Both need deliberate sequencing.

Phase one is capture. Every activity you intend to manage must land in the system without depending on a rep's discipline. Email and calendar sync for meetings, conversation intelligence for call recording and outcome tagging, automatic logging for touches. Anything that requires manual entry will be entered inconsistently, and inconsistent activity data is worse than absent activity data because managers act on it before discovering it's wrong. If a metric can't be captured automatically, either find a proxy that can be, or accept that it belongs in a qualitative one-on-one rather than on a dashboard.

Phase two is definition. This is where most rollouts quietly break. "Meeting held" needs a written definition — does a fifteen-minute rescheduling call count? Does an internal sync? Does a no-show that the prospect rebooked? Publish the definitions, encode them in the query layer, and version them. When a definition changes, annotate the dashboard, because an unexplained step-change in an activity trend destroys manager trust faster than a bad number does.

Phase three is the cascade. Wire each managed activity to the objective it is theorized to move, and each objective to its result. Write the theory down explicitly: "QBRs completed drives retention rate drives recurring revenue." Then test it. Some of your assumed linkages will not hold, and finding a dead linkage is valuable — it means you were coaching to an activity that doesn't matter, and you can stop.

Phase four is cadence. The framework only becomes management practice through a repeating rhythm: a weekly one-on-one anchored on activity gaps rather than deal status, a monthly review of objectives and process health, a quarterly reset of territory and account coverage. Separate the forecast call from the coaching call. When they're the same meeting, forecast always wins, because it's the one with an executive audience.

The two neglected layers. Account management and territory management are the least instrumented processes in most CRM deployments, and they're where the largest untapped leverage sits. For accounts, the useful construct is a health signal built from activity frequency, breadth of relationship, support burden, product usage where you have it, and time-to-renewal — not revenue, which is the lagging result you're trying to protect. For territory, the coverage questions are simple and rarely answered: what percentage of defined target accounts received any touch this quarter, how many touches did the top decile receive versus the bottom, and how much rep time went to accounts with no realistic potential.

Territory planning loses to deal-chasing because it never feels urgent. It is nonetheless the process that determines whether the sales force is fishing in the right pond at all — the highest-leverage decision in the system, made once a year, usually by whoever has time. A RevOps team that builds a defensible territory model balancing account potential, rep capacity, and coverage cost, and then *monitors adherence to the plan*, is applying this book at the layer where it pays best.

Adjacent applications. The hierarchy travels well outside the sales org, which is worth knowing because it makes cross-functional alignment cheaper. Customer success runs the same shape: NRR is a result, adoption depth and health scores are objectives, and structured check-ins and enablement sessions are the controllable activities. Marketing runs it too: pipeline sourced is the result, MQL-to-SQL conversion is the objective, and content production and campaign cadence are the activities. Building all three functions on the same three-tier vocabulary means a QBR can compare apples to apples instead of each team defending a bespoke scoreboard.

What holds up and what the book couldn't anticipate

What holds up. The three-tier separation is the durable contribution and has become invisible infrastructure — most people using leading-versus-lagging indicator language are running a compressed version of this framework without attribution. The four-process map remains a clean diagnostic for "what kind of sales force am I actually running." The manageability test is still the fastest way to kill a bad metric in a dashboard review. And the core prescription — manage activities, direct objectives, monitor results — remains the difference between a dashboard that changes behavior and one that just reports.

What has aged. The book was written when activity data had to be typed in by a rep, which shaped its assumptions in two ways. First, it necessarily favors coarse, countable activities, because those were the only ones you could reliably capture. Automatic capture and conversation analysis now make *quality* attributes of an activity measurable — whether discovery questions were asked, whether a competitor came up, whether the economic buyer was on the call — which is a richer coaching surface than the book could assume. Second, the book's implicit unit of work is a human touch. Modern motions include automated sequences, product-led signals, and self-serve behavior that produce buying progress with no rep activity attached. The hierarchy still holds, but the activity tier now has to include machine-executed and buyer-executed events, or your cascade will show mysterious results with no visible inputs.

A caution worth stating. Activity management degrades into activity surveillance if the metrics are used punitively rather than diagnostically. The purpose of counting proposals is to find the rep who stopped sending them and ask why — maybe their pipeline dried up upstream, maybe they're stuck on a skill. If the number is used as a stick, reps will optimize the number, and you'll get more proposals that shouldn't have been written. The book's framework assumes a coaching relationship; dropped into a low-trust culture, it produces measurable inputs and worse outcomes.

The RevOps takeaway in one line. Your job is not to report more, it is to arrange fewer metrics causally and make the controllable tier trustworthy. Everything else in the strategy — dashboard design, CRM schema, forecast methodology, comp design, coaching curriculum — falls out of getting that arrangement right.

Related questions

What is the single biggest mistake teams make applying this book?

Instrumenting too many activities at once. A forty-tile input dashboard fails identically to a results-only dashboard — no one knows which lever to pull. Pick two or three activities per process, prove the link to an objective, then expand.

Does the framework work for product-led or self-serve motions?

Yes, with an adjustment. The activity tier must include machine-executed sequences and buyer-executed product events, not just rep touches. Otherwise objectives move with no visible input and the cascade breaks, leaving you back at monitoring results.

How is this different from generic leading-indicator advice?

Leading-indicator advice lumps everything upstream of revenue into one bucket. Jordan and Vazzana split it into objectives you can only influence and activities you can control — the distinction that explains why most leading-indicator programs stall.

Which of the four sales processes should a startup instrument first?

Whichever carries your revenue. Short-cycle inbound motions start with call management; long-cycle enterprise starts with opportunity management; expansion-led businesses start with account management. Territory management is a later layer, added when coverage becomes the binding constraint.

Can this framework be used outside sales?

Readily. Customer success maps NRR to adoption objectives to structured check-in activities; marketing maps sourced pipeline to conversion objectives to campaign activities. Shared three-tier vocabulary across functions makes cross-team reviews comparable instead of each team defending a bespoke scoreboard.

FAQ

What is the central argument of Cracking the Sales Management Code?

That sales organizations track far more metrics than they can act on, and that the fix is separating metrics into three grades by manageability. Business results like revenue can only be monitored. Sales objectives like retention or share of wallet can be influenced through strategic direction. Sales activities like calls, meetings, and proposals can be directly managed through coaching and process. Effective management concentrates on the bottom two tiers, and the results follow.

How do I tell whether a metric is an activity or an objective?

Apply the manageability test: can you ask a rep to do more of it tomorrow, and would compliance be entirely within their control? "Send five more proposals this week" passes — it's an activity. "Raise your win rate" or "improve retention" fails, because the outcome depends on the buyer. Those are objectives you influence by directing effort toward the activities that produce them.

What are the four core sales processes, and do I need all of them?

Call management, opportunity management, account management, and territory management. You do not need to instrument all four. Different motions run on different processes — transactional inside sales lives on call and early-stage opportunity management, while enterprise field sales lives on account and territory management. Identify which process actually carries your revenue and build there first.

What does this mean concretely for a RevOps team?

It converts dashboard design from a collection exercise into a causal one. Rather than surfacing every available field, you build explicit activity-to-objective-to-result chains, capture activity data automatically so managers trust it, publish versioned metric definitions, and prune activities that show no measured relationship to the objectives they were supposed to move. The output is a shorter metric set arranged so a manager knows what to do, not just what happened.

Does the book give specific activity targets to hit?

No, and deliberately so. The authors provide the logic for selecting metrics rather than universal numbers, because the right activity volume varies enormously by industry, deal size, and cycle length. The recommended approach is deriving your own floors from your top-quartile performers' actual behavior over a trailing period and re-deriving them as the motion changes. Borrowed benchmarks carry false authority.

Is the framework still relevant now that AI captures and analyzes activity automatically?

More relevant, with one caveat. Automatic capture removes the historical failure point — reps under-logging until the dashboard became untrustworthy — and adds quality attributes the book couldn't measure, like whether discovery questions were asked. The caveat is that the activity tier must now include automated sequences and buyer-side product events, not just human touches, or the cascade will show objectives moving with no visible cause.

Sources

• https://www.mheducation.com/ — McGraw-Hill, publisher of Cracking the Sales Management Code
• https://www.vantagepointperformance.com/ — Vantage Point Performance, the authors' sales management research and training firm
• https://salesmanagement.org/ — Sales Management Association, research on sales management practice and metrics
• https://hbr.org/topic/subject/sales — Harvard Business Review, sales and sales management research
• https://www.gartner.com/en/sales — Gartner sales practice research on pipeline, forecasting, and seller productivity
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights — McKinsey growth, marketing and sales insights
• https://www.salesforce.com/resources/research-reports/state-of-sales/ — Salesforce State of Sales research on seller time allocation and CRM usage
• https://www.gong.io/resources/ — Gong research library on conversation and activity data in sales
• https://www.worldcat.org/ — WorldCat, library catalog for verifying the book's editions and publication data

Related on PULSE

• [Sales Management. Simplified. by Mike Weinberg — Cliff Notes Summary](/knowledge/bs0229)
• [High Output Management by Andy Grove — Cliff Notes Summary](/knowledge/bs0221)
• [From Impossible to Inevitable by Aaron Ross and Jason Lemkin — Cliff Notes Summary](/knowledge/bs0053)
• [Rework by Jason Fried and David Heinemeier Hansson — Cliff Notes Summary](/knowledge/bs0317)
• [Way of the Wolf by Jordan Belfort — Cliff Notes Summary & Key Takeaways](/knowledge/bs0020)