Measuring what a writer produces and what the work is worth

Productivity measures describe the quantity, quality and time involved in writing work. Value-added measures examine benefits to authors, readers and organisations, whose outcomes can depend on more than the writing alone.

Work volume, characteristics and time

Productivity can be measured through inputs such as manuscript pages or authors requesting support, outputs such as slide sets or people trained, and characteristics of the material handled. ¹ In a five-year editing-service study, original-research manuscripts had a median length of 12 double-spaced pages from cover page to conclusions, with half between 10 and 16 pages. ¹ For that service, manuscripts below about 10 pages raised concerns about underreporting or reluctant writing, while those above about 16 were candidates for shortening. ¹

Stopwatch-recorded editing time for original articles had a median of five hours, with half between 3.5 and 7.5 hours, at an overall rate of approximately four pages per stopwatch hour. ¹ That service could estimate total editing time from page count, while activity records by day, week or month showed where effort was requested, spent or under-resourced. ¹ Combining records from people in comparable jobs can establish ranges, with variation potentially more useful than average lengths, durations or page counts. ¹

Information work is difficult to measure and price, particularly insight, clarity, critical thinking and creative problem-solving. ¹ One approach is to ask what clients would pay for solving communication problems or feeling confident about publication quality, while explaining the service’s contribution beyond correcting spelling and punctuation. ¹ The relationship between effort and financial outcome can be uneven: one grants officer spent three years and a dozen applications without securing funding for newborn hearing-screening devices, while a research summary written in under an hour directly prompted a $2 million alumni gift. ¹

Quality and productivity measures in industry

An AMWA/CMR survey of pharmaceutical-company medical-communication managers found that all participating organisations used quality and productivity metrics. ^{2, 3} A preliminary report published in 2010 covering 13 companies and 15 respondents found completed documents per timeframe used by 87% and time to a specific draft by 80%. ⁴ It reported errors requiring correction at 67% and errors arising after publication at 53%. ⁴ The 2011 report gave 67% for errors or misinterpretations needing correction, 50% for published errata and 40% for meeting the document’s objective, such as regulatory approval or manuscript acceptance. ⁵

Quality measures used by at least a third of respondents included the following. ⁶

• Errors or misinterpretations requiring correction.
• Errata after publication.
• Regulatory requests for clarification or further information.
• Successful regulatory applications.
• Proportion of manuscripts accepted.
• Rejections per manuscript relative to first acceptances.

⁶

Productivity measures used by at least a third included completed documents or equivalents per timeframe, time to a specific draft, year-on-year comparisons and the proportion completed on schedule. ^{3, 7} Error and document counts had been used for decades; their popularity may reflect ease and apparent objectivity rather than validity. ³ Approximately a third used other quality measures and half other productivity measures, suggesting scope to identify additional useful metrics. ³ These findings came from a small sample focused on large companies and recruited through AMWA and existing research-company clients, limiting generalisability. ³

Costs, benefits and attribution

Adding value usually means reducing time, money or effort, or increasing information, accuracy, insight or clarity. ⁸ Value depends on the assessor, so measuring a writer’s contribution may mean measuring investigator time or reader effort saved. ⁸ An illustration proposes that 30 minutes improving an explanation read by 20,000 dentists could save up to 10,000 reader hours, assuming readers would otherwise spend time deciphering it; many would instead abandon an unclear passage. ⁸

A finished document does not separate a strong author’s contribution from an editor’s, and funding also depends on the idea, its fit with the funder, competition and circumstances within the agency. ⁸ Poor writing may justify rejection, but good writing alone cannot compensate for a poor idea. ⁸ Estimating the cost of misdirected, unread, misunderstood or unattempted communication may also help assess value. ⁸

One Cleveland Clinic editing service collected positive comments and regularly reported department chairs’ use of the service to administrators. ⁸ Another case involved a medical-writing group disbanded after Revlon’s pharmaceutical division was sold, followed by years of outsourced regulatory writing when scientist-prepared applications continued to be rejected. ⁸ Reputation can also contribute value: an Annals of Internal Medicine editor associated Mayo Clinic submissions with careful editing, formatting, tables and figures. ⁸

Instruments and observed outcomes

A 2013 MD Anderson project aimed to develop a 14-item scale for identifying manuscript problems by type and severity, to support fees, training tools and demonstrating editorial value. ^{9, 10} Its developers reported knowing no effective biomedical-manuscript assessment instrument and polled 20 departmental editors, then created, tested and adjusted a prototype. ^{10, 11} Validation among departmental editors began in March 2013; the expectation of a valid instrument was prospective. ^{12, 13}

At the Texas Heart Institute, a surgical group began using editorial support in 2003; its chief surgeon’s annual publications, previously around five from 1983–2003, increased each year to 23 in 2009. ¹⁴ An informal comparison of nine service users with nine non-users found similar publication rates during the three years before arrival and roughly twice the non-users’ rate among users by the third year afterwards. ¹⁴ Authors had publications in both three-year periods and less than 10% overlap with other included authors; users accessed the service at least annually and non-users never did. ¹⁴ The small informal analysis could not rule out self-selection, and users’ publication rate had already been increasing before arrival, although it rose faster afterwards. ¹⁴

Operational records

Metrics are numerical information about a process or activity and can support planning, resourcing, timelines, budgets and timely answers about a writer’s or department’s experience. ^{15, 16} Relevant records include the following. ¹⁷

• Client, therapeutic area, study phase and authoring start and end dates.
• Document type, such as protocol, clinical study report, paediatric investigation plan, manuscript or abstract.
• Business enquiries, wins and losses.
• Headcount and extended absences, including reasons.
• Geographical workload distribution in global departments.
• Writers’ experience by document type, therapeutic area, phase and population.
• Utilisation: billable hours relative to total billable and non-billable hours.
• Profit or loss against budget, and reasons.
• Timelines met, improved on or overrun, and reasons.
• Lessons from completed work.
• Audit findings involving writing projects or the department.

¹⁷

Business wins, future work and extended absences can justify recruitment and predict resource needs; low utilisation may prompt business development and high utilisation support hiring. ¹⁷ Client names distinguish new from repeat work, and wins relative to enquiries show acquisition success. ¹⁸ Authoring dates support realistic estimates that reduce stress and avoid mid-project resets, while consistently beating budgets or deadlines provides evidence of success. ¹⁷ Comparing recorded hours and cost with estimates identifies profit, loss and recurring underestimates, such as insufficient time per patient narrative accumulating across many narratives. ^{18, 16}

Regional project counts can guide growth, while metrics support budgets, training, service gaps, business development, client relationship indicators and positive staff feedback. ^{17, 18} Repeated audit findings can guide process improvement, and useful metrics may change as a business expands or diversifies. ¹⁸

Collection and review

Collection should begin by identifying the information needed, choosing the least labour-intensive method and checking existing sources before creating new systems. ¹⁹ Spreadsheets can have filters, conditional formatting and totals prepared at the outset, with only necessary source information transferred from email or other records. ²⁰

Practical collection guidance includes the following. ²⁰

• Use existing company summaries of headcount, finances or timesheet hours.
• Develop spreadsheet skills for totals and conditional formatting.
• Use graphs to reveal trends.
• Build searchable databases or dashboards for prompt answers.
• Enter information little and often before details are forgotten.
• Collect only useful data and review fitness for purpose, especially when requesting figures from a team.
• Present accurate metrics to stakeholders.
• Follow data-protection principles and company legal policies; obtain appropriate approval for personal or confidential information and observe lawfulness, fairness, transparency, purpose limitation, minimisation, accuracy, storage limitation, integrity and confidentiality.

²⁰

Although collection adds work, its benefits can outweigh the effort and improve understanding of past, current and forthcoming business activity. ²¹