BrandX campaign effectiveness · Step 4 of 6
Fieldwork and data collection
Where the plan meets reality: quotas, quality control, and course corrections
Introduction
Real-world practice. Fieldwork is where the research plan meets real respondents. Managing it well means active quota management, response optimization, and continuous quality control, so the final data genuinely reflects consumer opinion.
Portfolio Simulation
This portfolio recreates the complete fieldwork process through a Python-based simulation that mirrors how a commercial campaign evaluation would typically be executed. The simulated workflow applies the same quota controls, field management principles, and quality assurance procedures commonly used in live online research.
Field Management Timeline (14 Days)
Real-world practice. The 14-day field window moves through four phases, soft launch, fixes, full deployment, and closure, each with its own milestones and quality checkpoints.
Portfolio Simulation
The simulator follows this same timeline, varying its parameters by phase to reproduce the natural progression of fieldwork.
| Day 1-2 | Day 3 | Day 4-12 | Day 13-14 |
Day 1-2: If conducted, the study would begin with a small soft launch (approximately 50 respondents) to validate survey programming, routing logic, interview length, and respondent experience before the main field opens.
Day 3: This day would be reserved for resolving any issues identified during the pilot before progressing to full deployment.
Day 4-12: These days would focus on full-scale respondent recruitment while quotas, quality metrics, and field progress are continuously monitored.
Day 13-14: The final days would focus on quota balancing, quality review, and preparing an analysis-ready dataset.
A disciplined soft launch catches programming issues before they can touch the data, which is why the schedule reserves Days 1-3 for piloting and fixes.
Quota Management Dashboard
Real-world practice. Quota dashboards, refreshed throughout the day, show exactly where recruitment stands and where to intervene to keep the sample demographically balanced.
Portfolio Simulation
The simulation applies the same quota logic commonly used in commercial online studies, producing respondent distributions that closely follow the planned sample design.
| Gender | Target | Current | Completion % | Action |
|---|---|---|---|---|
| Male | 490 | 412 | 84% | Active recruiting |
| Female | 490 | 440 | 90% | Active recruiting |
| Gender diverse / other | Accepted to total | 13 | – | No quota |
| Demographic Cell | Target | Current | Completion % | Action |
|---|---|---|---|---|
| Females 18-34 | 175 | 164 | 94% | On target |
| Females 35-54 | 170 | 169 | 99% | Nearly closed |
| Females 55+ | 155 | 107 | 69% | Active |
| Males 18-34 | 175 | 157 | 90% | On target |
| Males 35-54 | 170 | 157 | 92% | On target |
| Males 55+ | 155 | 98 | 63% | Priority recruiting |
| Group | Target | Current | Completion % | Action |
|---|---|---|---|---|
| Control Group | 500 | 435 | 87% | Active |
| Exposed Group | 500 | 430 | 86% | Active |
Illustrative dashboard snapshot (Day 12 of a simulated 14-day field). If this study were conducted, females aged 35-54 would likely be nearing quota completion, while both 55+ cells could remain behind target. In this example, males 55+ represent the priority recruitment cell at 63% completion, and the Day 11 incentive scenario has not yet fully influenced recruitment. The exposed and control groups remain broadly balanced throughout the field.
Illustrative field outcome. Under this simulated scenario, hard quotas (gender, age band, region, and study group) would ultimately close at their planned targets. Age-by-gender interlocking quotas would operate as soft quotas with a ±10% flexibility band, reflecting common commercial practice for maintaining recruitment flexibility. In this example, every interlocking cell finishes within that tolerance, with the males 55+ segment reaching 104% of its soft target following the incentive scenario. Gender-diverse respondents (n = 16) contribute to overall gender and study-group quotas but are not allocated to the male/female interlocking cells.
In this simulated scenario, all hard quotas close at their planned targets before field closure, despite slower recruitment among older males.
Field Monitoring Dashboard
Real-world practice. With quotas set, daily tracking of the invitation-to-complete funnel drives real-time recruitment adjustments and keeps the exposed and control groups in balance. When participation or completion rates deviate from target, interventions follow.
Portfolio Simulation
The simulation reproduces realistic completion patterns commonly observed during commercial online fieldwork. The interactive chart below illustrates how recruitment could progress across different respondent segments.
Field productivity remains healthy across the simulated field, with over 90% of completed interviews passing all quality checks.
Daily Completions by Group
Filter Response Data:
The simulated data deliberately embeds three classic field scenarios, each visible in the chart: the Day 4 supplier boost, the Day 9 reminder recovery, and the late surge among 55+ respondents.
Field Intervention Scenarios (built into the simulated data):
Field Outcomes
Real-world practice. Field closure is judged on hard numbers: completion rates, data quality scores, and sample representativeness against plan.
Portfolio Simulation
The simulation produces an illustrative 1,295-record field file containing screen-outs, partial interviews, completed interviews, and quality failures, closely reflecting the structure of a commercial online panel dataset. Under this simulated workflow, quality assurance reduces the file to an analysis-ready sample of exactly 1,000 respondents.
Completion Details by Group & Device
Group Distribution
Device Distribution
The simulated field closes on time and on target: a clean 50/50 exposed-versus-control split and a device mix typical of modern online panels.
Field Management Playbook
Real-world practice. No field runs exactly to plan. Good field management means catching problems early and responding before they touch the data. This playbook lists the issues I anticipate at each phase and the response I keep ready.
Portfolio Simulation
The synthetic dataset does not encounter these issues and no incidents are claimed; the playbook documents the contingency planning real fieldwork demands.
Soft Launch: pilot-phase issues (5)
Full Launch: in-field issues (7)
Quality Assurance & Cleaning
Real-world practice. Layered quality control runs throughout the field, with automated checks flagging speeders, straight-liners, and other problem responses for review and removal.
Portfolio Simulation
The simulated field file deliberately includes the full texture of real fieldwork: screen-outs at A1 and A6, mid-survey abandons, invalid records, speeders, and straight-liners. This injected noise lets me demonstrate the complete, industry-standard exclusion and cleaning workflow.
Real-time Quality Monitoring
Speeder Detection
Threshold: ≈3.6 mins (vs 12.6 median)
Removed: 75 (6.8% of completes)
Termination Handling
Screen-outs, abandons, invalid
Excluded: 200 (15.4% of file)
Straight-lining
Flat C5 attitude grids
Removed: 20 (1.8% of completes)
Interview Length
Median: 12.6 minutes
Within the 12-minute design target
Data Cleaning Results: Under this illustrative scenario, the simulated field produces a dataset containing 1,295 records. After applying the quality assurance procedures described above, excluding 200 non-completes and removing 95 quality failures (8.7% of completes: 75 speeders and 20 straight-liners), the resulting analysis dataset contains exactly 1,000 high-quality respondent records. This removal rate sits within typical commercial norms of 5-10% of completes, and the identical figures are reproduced independently in the Step 5 preparation pipeline.
The final dataset meets all predefined quality standards and is ready for statistical analysis in Step 5.