Research outputs

The published research outputs of Slop University.

All research outputs of the University are published open access under CC BY 4.0, with persistent identifiers minted by the Office of Research Outputs.

Published outputs

Impact report

The Certificate and the Desk

Ingeborg Nwachukwu, Thandiwe Solberg, Torun Ezeigwe, Bram Ntuli · August 2026 · doi:10.5555/slop.yk6ulh

A School of Continuous Improvement Impact Report follows five years of Capability Sprint cohorts across four strands, comparing what graduates reported changing within 30 days of completion against what a twelve-month follow-up interview could still confirm, and finds the redesigned meeting formats, dashboard habits, and indicator retirements holding in some partner sectors and quietly reverting in others.

Paper

The Cursor Kept the Minutes

Verity Marris, Anouk Mensah, Casimir Beng · August 2026 · doi:10.5555/slop.900qei

A paper from the Schools of Emergent Priorities and Continuous Improvement audits forty call-centre floors against a four-stage maturity model of agent activity tracking, from a rostered break sheet to a continuous idle-seconds score. Trailing escalation and written-complaint rates show no association with maturity stage, and an ablation removing the score's cadence and idle-seconds components leaves that null result unchanged; maturity stage does predict a floor's odds of receiving its operator's quarterly commendation, rising to 9.7 times the odds of the least-tracked floors at the most granular stage.

Paper

The Roster Decided Who Got Called

Joost Nwosu, Solveig Adeyemi, Bram Ntuli · August 2026 · doi:10.5555/slop.q3r0tp

A paper from the School of Continuous Improvement exploits uneven casual-staff coverage across a nine-branch public library network to test whether a re-engagement call changes a flagged member's odds of reactivating. Across 3,140 members flagged by a card-tap engagement score over eighteen months, called and not-called members reactivate within sixty days at statistically indistinguishable rates, a null that survives a branch-level exogeneity check and a recency-only ablation of the underlying score.

Paper

The Wall Was the Last to Know

Torun Ezeigwe, Marit Osayande, Solveig Adeyemi · August 2026 · doi:10.5555/slop.likoy8

A paper from the School of Continuous Improvement treats a childcare operator's staggered rollout of a five-business-day photo-board update mandate as a quasi-experiment across twenty-four centres. Board-update lag falls from a mean of 11.4 to 4.6 days at mandate centres against no change among comparison centres, while drop-off recognition accuracy and handover-note accuracy move by an amount indistinguishable from the comparison group's ordinary drift; a routine walkthrough's own currency checklist item, by contrast, moves 44.1 points.

Paper

The Name on the Concrete

Ingrid Vasseur, Kwame Lindqvist, Lindiwe Achterberg · August 2026 · doi:10.5555/slop.j9cyre

A paper from the School of Emergent Priorities classifies 163 car-space ownership transfers across 21 strata schemes on a two-axis maturity model: how quickly a bay's stencilled name is repainted to match its new owner against how quickly the transfer is formally minuted by committee. The two measures order transfers close to independently of one another, and the dominant quadrant is one where the stencil has already caught up and the minute book has not, while each scheme's self-reported signage-compliance figure --- drawn from the slower record --- consistently overstates a directly audited rate.

Paper

The Countdown Decided First

Fenna Okoro, Verity Marris, Marek Solheim · August 2026 · doi:10.5555/slop.wd1e8f

A paper from the School of Emergent Priorities surveys 63 food-truck operators on a regional circuit about their trust in their ordering app's live queue-length countdown, cross-referenced against a fourteen-week season of the platform's own event log. Self-reported trust in the countdown as an early-warning signal shows no association with a truck's own season hit-rate: across 1,842 threshold crossings, a crossing preceded an actual stockout or walked-off customer within twenty minutes for only 9.0% of crossings on average, and trust does not track which trucks the warning serves well (r = 0.04). An ablation splitting the sample by the platform's original and mid-season-refreshed countdown display leaves operators' reported response mix unchanged. Reported response tracks the countdown's displayed minutes-remaining figure more closely than operators' own headcount of the physical line (r = 0.71 versus r = 0.18).

Paper

The Terminal Kept the Score

Mirela Hanke, Torun Ezeigwe, Marek Solheim · August 2026 · doi:10.5555/slop.27mugw

A paper from the School of Continuous Improvement validates a cafe franchise's point-of-sale "friendliness score" --- the seconds a terminal's default tip-suggestion screen holds a customer's attention, folded into quarterly outlet-manager reviews --- against realised tip percentage across 214,612 transactions and 44 outlets over five months. Prompt dwell time shows a negligible, non-significant association with realised tips (r = 0.06), a null that survives an ablation removing terminal hardware generation as a covariate, while the franchise's existing quarterly service-audit rating retains a moderate positive correlation in every region (r = 0.27-0.36). Dwell time instead varies threefold across the network's three terminal hardware generations, tracking equipment age rather than staff behaviour.

Paper

The Click Was the Whole Review

Dagny Okafor, Sten Okwuosa, Bram Ntuli · August 2026 · doi:10.5555/slop.05k152

A paper from the School of Continuous Improvement pairs interviews with 22 line managers against click-timestamp logs of 6,140 expense-reimbursement claims across 96 managers and four employers. Approval latency scales with claimed value at an identical rate regardless of a claim's eventual audit outcome, but the fastest-approved quarter of audited claims accounts for 34% of the mismatches a later reconciliation catches against a 25% baseline, and hiding itemised line detail from a subset of managers for a quarter leaves their mismatch-detection rate unchanged.

Paper

The Scanner Read the Stairs as a Drop

Torun Ezeigwe, Solveig Adeyemi, Joost Nwosu · August 2026 · doi:10.5555/slop.lkq5wu

A paper from the School of Continuous Improvement hand-codes 4,180 fall-detection alerts logged by a parcel courier's handheld scanner across nine depots into a five-category taxonomy, cross-referenced against the network's damaged-parcel claims register. Genuine-drop alerts are matched to a confirmed claim at 44.6%, an order of magnitude above every other category, while the network's own gentle-handling score treats all five categories identically; a firmware recalibration aimed at the two largest false-positive categories leaves the score's spread across drivers statistically unchanged.

Paper

The Sticker Knew the Date

Ronja Oyelaran, Petra Umbile, Iben Chikere · August 2026 · doi:10.5555/slop.edzmd1

A paper from the School of Emergent Priorities links a facilities portfolio's lift inspection-certificate registry, maintenance-ticket log, and an independently commissioned condition audit across 212 lifts and 74 buildings over four years. Certificate currency shows almost no association with recorded breakdown or entrapment events, while an independent condition-audit score and each lift's own door-cycle volume track both outcomes, and the two compliance rankings order the portfolio's buildings close to independently of one another.

Paper

Tightening the Threshold

Casimir Beng, Osei Vandermeer, Mirela Hanke · August 2026 · doi:10.5555/slop.wznk2x

A paper from the School of Continuous Improvement treats a supermarket chain's staggered, nine-wave rollout of a stricter self-checkout theft-flagging threshold as a natural experiment, finding no measurable change in shrinkage alongside an honest-shopper false-stop rate that rises in every wave.

Paper

Why We Turn It Off

Torun Ezeigwe, Solveig Adeyemi, Marit Osayande · August 2026 · doi:10.5555/slop.jga0gs

A paper from the School of Continuous Improvement interviews 47 staff across twelve break-room sites of a single facilities provider, finding that staff route around an auto-reordering supply shelf's own accounting once its weekly consumption report begins setting a departmental stationery budget. A parallel firmware recalibration aimed at the sensor's false-trigger rate leaves the frequency of logged staff overrides unchanged.