Content Operations Overhaul
The team wasn't the problem. The system was.
When declining quality put an entire writing team at risk, I showed leadership that replacing people would make things worse, then redesigned the system they were working in. Client churn stopped, quality non-compliance fell 63% and throughput rose 41%.
−33%
PRODUCTION HOURS
ROLE
Content Operations Manager & Chief Editor
TIMELINE
2016. Two-month redesign, with results measured over the following three months.
TEAM
7 Writers, 3 Editors, Leadership
TOOLS
Google Sheets
01
The challenge
Leadership wanted to replace the team.
Quality was declining, production costs were rising, and more than half of the agency's clients were threatening to leave.
Leadership believed the distributed writing team was underperforming and was preparing to replace it. Before touching the process, I built a risk-impact analysis showing what that would cost: a production gap while new writers ramped up, landing on clients who were already halfway out the door. Replacing the team wouldn't stop the churn. It would speed it up. That analysis earned me a mandate to audit the operation instead.
What the audit had to explain was a production operation with no consistent way to define quality, distinguish between types of editing, or check whether the work assigned matched the time available. Writers and editors worked across time zones. Editors skipped steps to meet deadlines, while repeated revisions and clarification requests consumed hours nobody had planned for.
Replacing people might have changed who did the work. It wouldn't have changed how the work was organized. I had two months to investigate and redesign the process without stopping production.
02
THE INVESTIGATION
The numbers challenged the original diagnosis.
I audited the production workflow and found a major gap between leadership's assumptions and the team's actual workload.
Quality assurance was 55.7% over budget, and roughly 70% of recorded production activity was non-productive: meetings, training and clarifications.
The production model also underestimated how much editing the work needed:
Edit flow | Planned | Actual |
|---|---|---|
F1 (Lightest) | 80% | 5% |
F2 | 15% | 25% |
F3 | 4% | 50% |
F4 (Heaviest) | 1% | 20% |
Leadership expected 95% of submissions to need light editing. In reality, 70% needed the two heaviest editing flows. The team was being measured against a production model that didn't match reality. So instead of replacing staff, I focused on defining editorial standards, measuring work consistently and redesigning the workflow around actual demand.
03
THE INSIGHT
You can't manage quality against a standard that doesn't exist.
The audit revealed that editors and writers didn't share a consistent definition of what different levels of editing required.
Proofreading, copy editing, and content editing were treated as loosely understood activities rather than distinct scopes of work. That ambiguity affected everything downstream: estimates, feedback, revisions, training, and accountability.
The solution wasn't simply to tell people to work faster or improve quality. I needed to make the quality standard explicit, measurable, and usable by everyone involved.
You can't blame a team for missing a bar that was never written down.
The breakthrough was to turn subjective editorial expectations into a shared operational framework.
Once the team could distinguish between edit types, assess work against consistent criteria, and identify why a submission exceeded its budget, quality became something the system could support rather than something individuals were expected to interpret independently.
04
THE SYSTEM
One editorial operating model connecting quality, cost, and production.
I redesigned the production workflow around four connected components:
Editing taxonomy: Defined proofreading, copy editing, and content editing to establish shared expectations.
Weighted scoring: Accounted for edit type, frequency, and word count to measure work consistently.
Assessment tool: Built a Google Sheets tool that calculated scores and flagged submissions as under, within, or over budget.
Production lifecycle: Reduced unnecessary meetings and clarifications while aligning workflows with actual editing demand.
Together, these components created a consistent system for defining quality, evaluating work, and managing production.
05
KEY DECISIONS
Design for accountability without turning measurement into punishment.
01
Define the work before measuring it.
I established consistent editing categories before introducing the scoring framework. Without shared definitions, a numerical score would simply give inconsistent judgments the appearance of objectivity. A metric can't fix an undefined standard.
02
Measure editing effort relative to submission size.
Raw edit counts weren't comparable across articles of different lengths or complexity. The weighted model accounted for edit types and word count, creating a more consistent basis for evaluating submissions and identifying work that exceeded budget. The score measured work against an editorial standard, not a person's worth.
03
Make the tool useful to writers, not just managers.
A scoring system introduced during a staffing crisis could easily feel like surveillance. I designed the taxonomy and assessment process around transparent criteria that writers could understand and use to defend their work. Results informed coaching and escalation rather than serving as an automatic disciplinary mechanism. Transparency was essential to adoption.
04
Redesign the workflow without stopping production.
The agency couldn't pause client delivery while I rebuilt its editorial operation. I introduced standards and process changes within the existing production cycle, prioritizing improvements that reduced repeated clarification and unnecessary coordination. The system had to work under real delivery constraints.
06
THE TRANSFORMATION
From subjective feedback to a shared production standard.
Before the redesign, writers and editors interpreted quality differently. Production estimates assumed light editing, while actual submissions required much heavier intervention. Revisions and clarification loops absorbed time without consistently improving the process.
After implementation, the team worked from defined edit categories, measurable thresholds, and a repeatable assessment workflow.
Editors could explain why work required additional effort. Writers had clearer expectations. Managers could distinguish between quality problems, process problems, and unrealistic production assumptions.
The system changed the conversation from “Who's failing?” to “What's happening in the workflow, and how do we improve it?”
BEFORE
Subjective production
Unclear standards
Inconsistent edits
Repeated clarification
Unpredictable cost
AFTER
Measured production
Defined standards
Consistent scoring
Targeted feedback
Predictable workflow
07
the outcome
The clients stayed. The team stayed. The work got better.
The churn stopped. The clients who had been threatening to leave stayed, and the team leadership had planned to replace kept its jobs.
Within three months of implementation, the redesigned system produced measurable improvements across quality and production:
Quality non-compliance fell 63%.
Throughput rose 41%.
Monthly production hours dropped from 1,116 to 743, as average time per submission fell from 6.2 to 4.1 hours.
Time lost to training and clarification per submission fell from 1.7 hours to 0.2, an 88% reduction.
The same team produced better work, faster, once the system gave it clear standards and removed unnecessary work. The figures reflect the redesign as a whole rather than isolating each individual change.
08
What I Actually Designed
The system behind consistent editorial production.
The visible output was a scoring tool and a revised workflow. The underlying work was designing an editorial operating model.
I built the risk-impact analysis that changed leadership's decision and won the mandate for the audit.
I defined the quality standard: a three-tier editing taxonomy with error codes for each tier.
I translated it into a weighted measurement framework and built the Google Sheets tool that made it usable day to day.
I redesigned the production lifecycle around actual workload, with explicit ownership and escalation paths.
I built the capacity and staffing model that matched assignments to the time the work actually took.
I created the style guide and a topic development training deck so standards held as new writers joined.
I also handled the human side of implementation: making the criteria transparent, supporting coaching, and introducing changes without disrupting client delivery.
The value wasn't making editors work faster. It was designing a system that made quality and efficiency achievable together.
09
THE BIGGER IDEA
Fix the system before blaming the people.
When performance declines, replacing staff can look like the most direct response. But without understanding how work is defined, assigned, measured and reviewed, a new team inherits the same problems.
This project showed the value of investigating the operating model before making staffing decisions. By making standards explicit and redesigning the workflow around actual demand, I created the conditions for the existing team to produce better work more efficiently.
A new team would have walked into the same broken system.
10
Reflection
Measurement only works when people trust the standard behind it.
The most consequential decision was introducing a scoring system without turning it into a punitive performance tool.
A number can create clarity, but it can also create false certainty if the underlying definitions are inconsistent or people don't understand how it's calculated.
I started by defining the work, then built the measurement model and made its criteria transparent. That gave writers a clearer basis for understanding feedback and gave editors a more consistent way to explain their decisions.
The broader lesson was that operational improvement requires both system design and change management. A workflow can be mathematically efficient and still fail if the people using it don't understand or trust it.
If I revisited the project, I'd formalize ongoing calibration between editors and review the scoring thresholds periodically as content complexity and client expectations changed.
Want to see the actual error taxonomy, or how the escalation paths worked in practice? Reach out and I can go into the details!