03 · What You Need to Know
How to Build a Realistic Recruitment and Data Collection Timeline
Separate start-up, recruitment, and data collection
Researchers sometimes use "data collection period" to describe everything between proposal approval and analysis. That can conceal several different stages.
Start-up includes whatever must happen before recruitment or data collection can legitimately begin. Depending on the study, this may include ethics or institutional approval, site permission, contracts or agreements, preparation of recruitment materials, instrument programming, equipment setup, staff training, pilot testing, database preparation, and scheduling.
Recruitment involves identifying, approaching, screening, consenting, and enrolling eligible participants according to the approved procedures.
Data collection involves obtaining the observations required by the study. Depending on the design, this may occur immediately after enrollment or continue through interviews, surveys, observations, assessments, repeated measurements, interventions, or follow-up visits.
Recruitment period
The time required to identify, approach, screen, and enroll the participants needed by the study.
Data collection period
The time required to obtain the measurements, responses, observations, records, or follow-up information required from the enrolled sample or other data sources.
These periods may overlap substantially, but they are not necessarily identical.
The clock should not start before the study can actually start
If your proposal allocates June and July for data collection but site authorization is not expected until late June, you do not really have two months.
Start-up activities can consume a meaningful portion of the research schedule. The National Institute of Mental Health advises investigators preparing clinical research studies to consider start-up activities such as regulatory and ethics approvals, staffing, training, equipment, study materials, contracts, and other preparatory requirements when developing timelines.
The exact requirements vary considerably across research settings. A simple anonymous online survey may require relatively little operational setup once the applicable approvals are complete. A multisite intervention, clinical study, laboratory experiment, or project involving restricted populations can require considerably more.
Estimate start-up separately rather than quietly borrowing time from recruitment when approvals take longer than expected.
Start with the number of usable participants you actually need
Your timeline should be based on the sample required by the design, not merely the number of people you hope to invite.
If your analysis requires 250 usable cases, the relevant planning question is how long it will take to produce approximately 250 usable cases after accounting for eligibility, participation, incomplete data, and attrition where relevant.
That may require approaching substantially more than 250 people.
Before estimating the timeline, therefore, establish whether you can realistically recruit the participants your study requires and how large the recruitment pool must be.
Estimate a recruitment rate, not just a recruitment target
A target tells you where recruitment needs to end. A rate tells you how long getting there may take.
Suppose you need 240 completed participants. If your recruitment process produces approximately 20 usable participants each week, recruitment might require around 12 weeks under those assumptions. If it produces five per week, the same target requires approximately 48 weeks.
The population has not changed. The recruitment rate has.
Useful evidence for estimating a rate may come from previous studies at the same site, comparable research, pilot or feasibility work, administrative information about how many eligible people appear during a typical period, or preliminary discussions with experienced site personnel.
Avoid borrowing a recruitment rate from a superficially similar study without considering differences in eligibility criteria, participant burden, setting, recruitment strategy, and population.
Build the recruitment estimate from a funnel when necessary
When direct historical recruitment rates are unavailable, you can construct a planning estimate from the stages between initial access and usable data.
The percentages are planning assumptions, not promises. Their usefulness depends on the evidence supporting them. The calculation is valuable because it exposes what your timeline assumes about the recruitment process.
Recruitment rates may not remain constant
A simple estimate often assumes the same number of participants will enroll every week or month. Real recruitment can be less tidy.
The first weeks may be slow while staff learn procedures. Recruitment may accelerate once additional sites open. A large initial response to an announcement may be followed by a sharp decline. Academic calendars, holidays, seasonal patterns, clinical caseloads, staff availability, competing studies, and community events may alter the pool over time.
For this reason, a single average rate should be interpreted cautiously. Where possible, examine whether the population or recruitment setting has predictable periods of higher or lower availability.
A study recruiting university students, for example, may encounter a very different population during examinations or semester breaks than during ordinary teaching weeks.
Site capacity can limit recruitment even when participants are plentiful
Imagine that hundreds of eligible participants are available, but your study requires a 90-minute laboratory session and the laboratory can accommodate only two participants per day.
Your bottleneck is no longer participant availability. It is operational capacity.
Similar constraints arise when one interviewer conducts all interviews, one clinician performs all assessments, one piece of equipment is shared among several studies, or data collection can occur only during limited site hours.
Estimate the maximum number of participants your research process can handle per day or week. Then compare that capacity with the recruitment target.
Recruiting participants faster than you can process them can create scheduling delays, participant inconvenience, staff overload, and potentially protocol deviations.
Participant burden affects both recruitment and scheduling
A five-minute online questionnaire and a three-visit experimental study should not receive the same recruitment assumptions simply because they seek the same number of participants.
Participation burden can influence willingness to enroll, appointment availability, cancellations, retention, and completion. Travel, childcare, work schedules, privacy concerns, physical procedures, lengthy interviews, repeated surveys, and technology requirements can all affect the timeline.
Look at the study from the participant's perspective. How easy is it to find a suitable appointment? How much notice is required? What happens when someone cancels? Can sessions be rescheduled? How many reminders or follow-up attempts are permitted and appropriate under the approved procedures?
These practical details determine how quickly enrollment turns into usable data.
Recruitment and data collection may overlap
In many studies, data collection begins before recruitment finishes.
For a cross-sectional survey, a participant may complete the questionnaire immediately after enrolling. For interviews, researchers may conduct early interviews while continuing to recruit additional participants. In an experiment, participants may be enrolled and tested throughout the recruitment period.
Overlap can shorten the total project duration, but only if your staffing and procedures can support both activities simultaneously.
If the same researcher must recruit participants, schedule sessions, conduct interviews, transcribe recordings, manage data, and continue coursework or clinical duties, theoretical overlap may create a workload bottleneck rather than a time saving.
Longitudinal studies have two different time problems
Longitudinal research requires attention to both the recruitment window and the follow-up duration.
Suppose each participant must be followed for six months. If the first participant enrolls in January and the final participant enrolls in April, data collection does not finish in June simply because six months have passed since recruitment began. The final participant's six-month measurement occurs in October.
This is a common source of unrealistic schedules. Researchers remember the six-month follow-up but forget that participants enter the study at different times.
Retention can extend the workload even when it does not extend the protocol
Repeated-measures and longitudinal studies also require active retention work.
Participants may need reminders, rescheduling, updated contact information, repeated invitations, or other approved retention procedures. Missed follow-ups can require additional contact attempts. Some participants will withdraw or become unreachable.
The National Institute of Mental Health recommends that recruitment and retention planning account for the particular population, study burden, site characteristics, staff, and strategies required to maintain participation.
Retention is therefore not merely a percentage deducted from your final sample. It is work that occupies staff time throughout data collection.
Qualitative recruitment also needs a timeline
Qualitative research does not escape recruitment planning simply because it may involve fewer participants.
Recruiting people with specific experiences can take considerable time, especially when the population is specialized, sensitive, or difficult to reach. Interviews themselves may require scheduling, rescheduling, travel, transcription, memo writing, and preliminary analysis.
Some qualitative designs also use iterative sampling in which emerging analysis influences whom researchers seek next. In such cases, it may be inappropriate to specify the exact final sample solely through a fixed numerical calculation at the beginning.
You should still estimate how quickly appropriate participants can be recruited, how much time each case requires, and whether the iterative process can fit within the available research period.
Do not forget nonparticipant data collection
Not every project recruits human participants, but similar timeline logic applies to other data sources.
Archival research may depend on archive opening hours, retrieval requests, digitization, travel, or limits on the number of materials that can be accessed each day. Laboratory research may depend on equipment throughput, specimen preparation, incubation, calibration, or repeated trials. Field observations may depend on events occurring at particular times. Secondary-data projects may require access approval and substantial data preparation before analysis.
The core question remains the same: what process produces one usable unit of evidence, how many such units do you need, and how quickly can the process realistically operate?
Include time for data checking while collection is still underway
Waiting until the end of data collection to inspect the data can allow preventable problems to continue for weeks or months.
Where appropriate and consistent with the protocol, check whether expected records are arriving, variables are being captured correctly, files are readable, equipment is functioning, interviews are being recorded, participant identifiers are consistent, and data are being stored as intended.
This does not mean repeatedly testing hypotheses until the results look satisfactory. It means monitoring data integrity and operational performance.
Discovering after 300 participants that a survey branch was programmed incorrectly is a particularly memorable way to learn the difference.
Add contingency time for plausible delays
A timeline built on everything happening exactly as planned is not a realistic timeline.
Contingency should reflect the vulnerabilities of the particular study rather than an arbitrary universal percentage. A straightforward online survey with several independent recruitment channels may need less schedule protection than a multisite study dependent on sequential approvals, specialized equipment, and repeated participant visits.
Ask which activities are outside your control. Site approvals, participant availability, external data extraction, equipment repair, organizational schedules, weather-dependent fieldwork, and third-party services may deserve additional allowance.
Contingency is not wasted time. It is protection against predictable uncertainty.
Do not use contingency to hide an already impossible schedule
Adding two extra weeks does not rescue a project whose expected recruitment process requires nine months when only four months are available.
First calculate the best evidence-based estimate for the work itself. Then add reasonable allowance for uncertainty. If the resulting timeline exceeds the deadline, the study needs redesign rather than more optimistic arithmetic.
You may need additional sites, a less restrictive population, more recruitment capacity, different data-collection procedures, a narrower research question, or another study altogether.
Watch Out
Do not set the recruitment period by subtracting analysis and writing time from your final deadline and then assume the required sample will somehow fit inside what remains. Estimate recruitment and data collection from the study's actual throughput first. If the resulting timeline does not fit, the design needs to change.
Set recruitment monitoring points before data collection begins
A timeline becomes more useful when it contains decision points rather than only start and end dates.
Suppose your plan requires an average of 25 usable participants per month. After two months, you have 18 in total rather than approximately 50. That difference deserves attention while there is still time to respond.
Predefined monitoring points can prompt you to examine whether recruitment channels are functioning, whether eligibility assumptions were wrong, whether additional approved sites or strategies are needed, or whether the project requires redesign.
A monitoring point should not become an excuse to manipulate recruitment in ways inconsistent with the protocol. Its purpose is to identify feasibility problems early.
Recruitment should stop according to the design, not simply when time expires
A deadline is not a sampling strategy.
For a quantitative study, stopping early because the semester ended may leave an inadequate sample for the intended analysis. For qualitative work, stopping solely because a convenient number has been reached may ignore the logic of the chosen methodology.
If the study cannot obtain sufficient evidence within the available recruitment period, acknowledge the feasibility problem and determine what methodological response is appropriate.
The alternative is allowing the calendar to silently rewrite the study after data collection has already begun.
