Velocity measures how long a marketing decision takes to produce market impact. Your org chart hides it because each team owns a task, while nobody owns the waiting between tasks. That elapsed time can be the most expensive number in marketing: spend continues, buyer intent cools, and the next useful decision arrives late.
A bigger team can still ship less
Our white paper, Right Strategy. Wrong Operations., compares Organization A, a $400 million B2B software company with 32 marketers, with Organization B, a $180 million company with 19. A shipped six campaigns a quarter at an average brief-to-launch cycle of 43 business days. B shipped 22 in nine business days. The campaign types were the same, with comparable complexity. The larger company lost on operations.
Treat that comparison as a reason to investigate, not a universal nine-day target. I have seen teams add capacity while leaving every approval intact. The extra people create more work for the same decision queue. A designer can finish on time, a channel manager can meet every deadline, and the campaign can still spend weeks between them. Department-level performance won't expose that gap.
Map one campaign before changing the org chart
Choose a recent, ordinary campaign. Bring its brief and task history into a working session with the people who handled it. Reconstruct the path from approval to launch. For each step, record when the work became ready, when someone started it, when they finished, and when the next person accepted it. Use task logs and email timestamps rather than remembered durations.
Count every handoff between owners and every approval gate. Mark revisions separately from approvals, including why work came back. Ask which decisions required judgment and which merely required permission. Ask who could have acted sooner, and what prevented them. If a ticket sat in a queue, find out whether the team lacked capacity or nobody knew it was theirs. Those problems need different fixes.
Suppose a campaign took 24 business days, with six days of active work and 18 of waiting, across nine handoffs and five approvals. This is a hypothetical diagnostic, not company data. Faster asset production won't remove the 18 days. Trace the longest waits first. Use the critical path when tasks overlap, so you don't add simultaneous delays together and claim savings you cannot deliver.
Start with a usable sample
You don't need a new dashboard to establish a baseline. Start with 20 consecutive records for each measure, using a longer period if volume is low. Include difficult cases rather than choosing clean ones. Separate comparable campaign types and lead intent levels. Reconcile timestamps to one time zone. Report missing records and sample size beside the result; an unknown timestamp is not zero elapsed time. If history is unreliable, log the next 20 prospectively.
Report the median and the 90th percentile, the point below which roughly nine out of ten observations fall. The median describes the usual experience; the slower tail exposes neglected work. Keep the definitions fixed. The first targets below are operating starting points I would propose after a baseline, not industry benchmarks.
Brief to launch measures the production queue
Start the clock when a complete brief is approved for execution. Stop when the program is live and usable by its intended audience. Pull approval from the project system's status history or the approval email. Confirm launch with the platform publish log and a live check. A scheduled email or a finished landing page does not establish that the whole program launched.
When records are messy, reconcile the sampled campaigns against launch calendars and asset histories. Keep incomplete briefs in a separate intake measure so teams cannot improve the number by postponing approval. Count business days using one calendar, and retain revision time in the cycle. For a repeatable campaign type, propose a 20% reduction in median cycle time over 90 days, with no increase in launch defects. Review the slowest cases weekly. A new product launch deserves a separate baseline.
Speed-to-lead starts when the buyer raises a hand
For a demo request, start with the form submission timestamp, not the time the CRM finally creates or assigns the record. Stop at the first human contact attempt recorded in the dialer or sales activity log. Define that endpoint explicitly; an automated acknowledgment doesn't count. Track successful two-way contact separately. A quick unanswered call and a conversation are different outcomes.
Join form submissions to CRM records using submission IDs or email and time. Audit sampled calls and emails if activity logging is inconsistent. Report both elapsed minutes and staffed-hours performance; otherwise overnight leads disappear behind a convenient clock. For high-intent requests, I would pilot a 15-minute first-attempt SLA during staffed hours. Publish the coverage window and an explicit next-shift deadline for requests outside it. Assign a backup when the primary owner is unavailable.
In a prior leadership role, I worked with a financial services firm spending $4.8 million a year on demand generation that was losing 40% of leads to routing delays. A 15-minute routing SLA took routing from more than 48 hours to under 15 minutes, and the sales-accepted lead rate rose 45%. This was before Nova Nexus. Routing is only assignment: measure the subsequent human response too, or you can celebrate a faster transfer into another unattended queue.
Time in stage exposes stalled progression
Use CRM stage history to subtract each stage-entry timestamp from its exit timestamp. Preserve repeat entries when opportunities move backward. A current stage field and a last-modified date cannot reconstruct that history. If history is missing, use dated activity notes for a labeled manual sample, then enable stage-change logging for future records. Define what qualifies a deal to enter and leave each stage before comparing teams.
Include the age of deals still open in the stage alongside completed transitions. Reporting only exits hides the deals that never move. Separate market segments and distinguish buyer-requested pauses from internal delays. A reasonable first target is a 20% reduction in the 90th-percentile duration of one internally constrained stage over 90 days, with conversion held steady. Moving deals forward without qualification improves the clock and damages the forecast.
Time to insight ends with a decision
Start with the verified launch timestamp. Stop at the first documented decision based on results, whether to continue, change or stop the program. Use the decision log or dated meeting notes, linked to the evidence reviewed. A dashboard refresh isn't a decision. Record who decided and what action followed. If the history is missing, inspect review notes for your sample; don't substitute the report delivery date.
Set the evidence threshold before launch. A low-volume enterprise program may need longer than a high-volume acquisition test. For recurring campaigns with sufficient early signal, propose a first decision within ten business days. Where the evidence takes longer, schedule the review for the first justified window and target a decision within two business days of reaching the threshold. Logging 'keep waiting' every week should not pass as useful insight.
Remove permission queues without removing controls
Delegate approval within written thresholds. A channel owner might approve work inside an agreed budget and established claims policy, while new regulated claims still require legal review. Name the approver and backup, define the response deadline, and establish escalation. Silence should never count as approval for regulated content. If every routine decision still reaches the CMO, the delegation hasn't happened.
Remove handoffs that transfer no useful judgment. Give one campaign owner responsibility through launch instead of moving coordination between departments. Use approved brief templates and reusable assets to prevent avoidable rework. A template needs a maintained owner and a version date; copying last year's approved language is not a control. Review recurring rejection reasons before asking the team to work faster.
For lead routing, specify assignment logic, acceptance deadlines and fallback ownership. Alert on breaches and test the fallback. Automating a broken rule sends leads to the wrong place faster. Run one fix on one workflow before expanding it, and check that shorter cycle times haven't increased rework or reduced conversion.
Put elapsed time on the leadership scorecard
Velocity belongs beside pipeline and spend. In the Nexus Performance Model, the four outcomes are Strategy & Operations, Success, Velocity and Efficiency. For each velocity measure, show the baseline, current median, 90th percentile, target and accountable owner. Include sample size and the share of records with usable timestamps. Attach a quality guardrail so nobody gets rewarded for shipping broken work or advancing unqualified deals.
Review exceptions weekly with operating owners; review the trend monthly with leadership. Give one executive authority to resolve cross-team delays, even where marketing shares the process with sales. When a target slips, ask which queue grew and which decision needs changing. Don't accept 'the team is busy' as the whole explanation. Capacity may be the cause, but you need evidence before adding headcount.
What to do Monday morning
Pick one recent campaign and reconstruct its path. Ask marketing operations for a sample of high-intent leads with submission, assignment and first-attempt timestamps. Name someone to maintain the four measures, then put the first baseline review on the calendar. Choose one avoidable wait to remove this month.
At Nova Nexus, that follows Insight. People. Results. Measure the actual operation before changing it, give people the decision rights to fix the delay, then check the outcome. The free Marketing Operations Scorecard can help start the discussion. Bring timestamps to the next leadership meeting. They will tell you something the org chart cannot.
See how your operation compares.
Score your marketing operating system in five minutes, or talk it through with our founder.
Get new Insights by email.
New papers and points of view for CMOs, about once a month. No other mail. Unsubscribe any time.

