Case file · Technology (software and platforms)

Larkspur Software

A Toronto scheduling app built to give shift workers control over their time is asked by its biggest clients to watch them instead.

Source: Simulated data Fictional organization Advanced Updated 3 November 2026

The decision

Should Larkspur build the worker-monitoring features its largest clients demand, and if not, what should it offer instead?

Larkspur's app lets people who work in shifts see their rotas, swap shifts and record their hours, and its clients pay for each active worker each month. Its three largest clients, logistics firms that provide more than a fifth of its revenue, want Pulse: productivity scores, idle-time alerts and rankings for every worker. They will move to a rival if Larkspur will not build it. Building it pays best, but it contradicts the company's mission, most of its engineers have signed a letter against it, and the workers who use the app say they do not want it.

Organization
Larkspur Software
Sector
Workforce scheduling software, sold as a subscription
Location
Toronto, Canada
Size
140 employees, of whom 58 are engineers; 990 client organizations
Legal form
Privately held company, owned by its founders, employees and a venture capital investor
Figures in
CAD (C$), thousands of Canadian dollars (C$000), and dollars for prices
Functions
Strategy and management, Operations management, Finance and accounts
For
IB DP SL, IB DP HL, Undergraduate, MBA
Key concepts
Ethics, Change
Status
Fictional case. An invented organization. Every figure is simulated for teaching and checked for consistency.

Section 01: The story so far

Priya Nandakumar spent eight years building nurses' rotas in a Toronto hospital, on spreadsheets that nobody liked. In 2018 she and Owen Kalu, a software engineer, founded Larkspur to build something better: an app that lets people who work in shifts see their schedules, swap shifts with colleagues and record their hours, and that lets managers fill gaps without phoning around. Its mission, printed on the wall of the office, is to give people who work in shifts more control over their time.

Larkspur now has 990 client organizations: care homes and clinics, shops and restaurants, and warehouses. They pay for each active worker each month, and the workers use the app free (Exhibit 2). Revenue grew by 20.0% in 2025, and Larkspur made its first profit (Exhibit 1). A venture capital investor owns 35% of the company and expects to lead a new funding round in 2027, if growth continues (Exhibit 9).

In the spring of 2026 Larkspur's three largest clients, all logistics firms, made a joint request. They want Pulse: an add-on that scores each warehouse worker's productivity hour by hour, alerts managers when a worker is idle, tracks where workers are during a shift and ranks them against one another. They will pay C$2.00 a worker a month for it, and they say that if Larkspur will not build it within a year they will move to a rival platform that will (Exhibit 7).

The request split the company. Thirty-four of the 58 engineers signed an open letter saying that Pulse would turn a tool for workers into a tool for watching them (Exhibit 8). A survey of the people who use the app found most of them uncomfortable with being ranked (Exhibit 5). The sales team warned that some care and retail clients chose Larkspur precisely because it is not a surveillance tool.

Priya has asked her team to cost four responses: build Pulse as the clients asked; build Pulse Lite, with team-level data, no rankings and individual data only for workers who choose to share it; open an interface that lets clients plug in their own analytics tools; or decline and win more care and retail clients instead (Exhibit 6). The board meets in June.

Section 02: The industry

How the industry works and what is changing in it, in general terms. The figures in this case are the business's own, and simulated.

Workforce management software helps employers with many hourly workers plan rotas, record hours and pay people correctly. It is usually sold as a subscription, priced for each user each month and run on the supplier's own servers. Revenue recurs, so keeping clients matters as much as winning them, and the extra cost of serving one more user is low, which is why gross margins in software tend to be high.

Apps that workers carry on their phones collect detailed data about when and where people work. Employers increasingly want to use such data to measure productivity, and suppliers compete to offer analytics. Workers, unions and regulators are increasingly concerned about monitoring at work: what is collected, whether people know, and how decisions about them are made.

A scheduling app needs workers to use it as well as employers to buy it. Its ratings in the app stores, and what workers say about it to one another, affect whether employers choose it. Large suppliers of human resources software offer scheduling as part of wider suites, so smaller specialists compete on how well they serve particular industries and the people who use them.

Section 03: The key challenge

Larkspur must answer its largest clients without losing either them or itself. Saying yes would keep more than a fifth of its revenue and add a profitable product, but it would make the company a supplier of the kind of monitoring its mission was meant to replace, and it risks its engineers, its reputation with workers and its other clients. Saying no protects all of that and turns a profit into a loss a year before the company needs investors' confidence. The board has to decide what Larkspur is for, and what that is worth.

Section 04: Financial information

Larkspur's revenue grew by 20.0% in 2025, to C$4,752k, and it earned its first profit before interest and tax, C$184k, a profit margin of 3.9%. Its gross profit margin is high, 75.0%, because serving one more worker costs little; most of its costs are engineers' salaries. The three logistics clients pay C$1,080k a year, 22.7% of revenue. If they left and nothing replaced them, Larkspur would lose three quarters of that in gross profit and return to a loss. It has C$2,900k of cash to pay for any development.

Exhibit 1

Summary statement of profit or loss, years to 31 December (C$000)

Source: Simulated data
Item 2024 2025
Revenue3,9604,752
Cost of sales1,0301,188
Gross profit2,9303,564
Expenses3,0203,380
Profit before interest and tax−90184

Cost of sales is hosting, customer support and payment fees, and is about a quarter of revenue. Expenses in 2025 were C$2,100k for product development, C$900k for sales and marketing and C$380k for administration. Larkspur has no borrowing, and C$2,900k of cash at the end of 2025.

Source Simulated teaching data, from Larkspur's accounts.
Exhibit 2

Clients and revenue by segment, 2025

Source: Simulated data
Segment Clients Active workers Price per worker per month (C$) Revenue (C$000)
Care and health38031,0004.501,674
Retail and hospitality52027,0004.501,458
Logistics and warehousing9037,5003.601,620
All clients99095,5004,752

Revenue is active workers times the price, times twelve months. Logistics clients pay less per worker because they buy in volume. Of the 37,500 logistics workers, 25,000 work for three clients: Northgate Logistics (12,500), Halvorsen Freight (7,500) and Bayline Distribution (5,000).

Source Simulated teaching data, from the billing system.
Exhibit 6

Four responses to the request, forecast effect from 2027 (C$000)

Source: Simulated data
Response Development in 2026 Change in revenue Change in gross profit Change in other costs Change in yearly profit before interest and tax Profit before interest and tax in 2027
A. Build Pulse as asked1,800588441200241425
B. Build Pulse Lite1,50032424315093277
C. Open an API for clients60096725022206
D. Decline and refocus0−864−648300−948−764

Changes are against 2025, at the 75% gross margin on any change in revenue; other costs are each response's own running costs. The estimates are the product and sales teams' judgements.

  • A. Pulse at C$2.00 a worker a month for the three clients' 25,000 workers and for half of the other 12,500 logistics workers (6,250); clients in care and retail with 3,000 workers leave.
  • B. Pulse Lite at C$1.50 a month for Northgate and Halvorsen (20,000 workers) and for 10,000 workers in care and retail; Bayline, with 5,000 workers, leaves.
  • C. The interface at C$0.50 a month for the three clients' 25,000 workers; clients in care and retail with 1,000 workers leave.
  • D. The three clients leave; an extra C$300k a year of sales and marketing wins clients in care and retail with 4,000 workers.

Development is paid once, in 2026, from cash.

Source Simulated teaching data, prepared by the product and sales teams.

Section 05: The rest of the evidence

6 exhibits. Each is labelled with what kind of evidence it is.

Exhibit 3

What Pulse would do

Source: Simulated data
  • Productivity scores. A score for each worker for each hour, from scans, picks and tasks completed.
  • Idle-time alerts. A message to the manager when a worker has done nothing measurable for a set number of minutes.
  • Location during the shift. Where in the building each worker is, from the phone's position.
  • Rankings. Each worker ranked against colleagues, shown to managers every day.
Source Simulated teaching data, from the clients' specification.
Exhibit 4

Pulse Lite and the interface, as the product team proposes them

Source: Simulated data
  • Pulse Lite. Productivity data for teams and sites, never ranked by worker. Each worker sees their own data and decides whether to share it with a manager. C$1.50 a worker a month.
  • The interface. A connection that lets a client send Larkspur's data to an analytics tool of its own choosing. Larkspur builds no monitoring itself. C$0.50 a worker a month.
Source Simulated teaching data, from the product team's proposal.
Exhibit 5

The people who use the app

Source: Simulated data
Rating in the app stores
4.6 out of 5
Workers who say Larkspur helps them get shifts that fit their lives
78%
Workers who would be uncomfortable if their employer used the app to rank them
81%
Workers who would consider leaving a job that tracked them minute by minute
46%

An in-app survey of 2,400 workers in April 2026. Those who answer surveys in an app they use may not represent all its users.

Source Simulated teaching data.
Exhibit 7

The clients' request

Source: Simulated data

Our margins depend on productivity, and we need data by worker, by hour, with alerts and rankings our managers can act on. Your scheduling is the best we have used. But if Larkspur will not build this within twelve months, we will move to a platform that will when our contracts come up for renewal.

Marcus Leclerc, chief operating officer, Northgate Logistics, writing for the three clients
Source Fictional letter, written for this case.
Exhibit 8

The engineers' letter

Source: Simulated data

We built Larkspur to give people who work in shifts more control over their time. Pulse would use the same app, on the same phones, to watch them minute by minute and rank them against each other. We do not think a company with our mission should build it, and some of us will not.

Open letter to the founders, signed by 34 of Larkspur's 58 engineers
Source Fictional letter, written for this case.
Exhibit 9

The investor's view

Source: Simulated data

We invested in Larkspur because workers love it, and that is still its advantage. But we need to see revenue growing by at least 20% a year before we lead the next round. Losing a fifth of revenue in one go would be very hard to explain.

Hannah Osei, partner, Lakeshore Ventures
Source Fictional email, written for this case.

Section 06: Constraints

  • The three clients' contracts come up for renewal in twelve months, and they have said they will not renew without worker-level monitoring.
  • Larkspur has C$2,900k of cash and no borrowing; any development must be paid for from it until the next funding round.
  • The investor expects revenue growth of at least 20% a year ahead of the round planned for 2027.
  • Larkspur's legal adviser has warned that clients must tell workers what is monitored and why, and that a supplier that designs the monitoring may share responsibility if the data is misused.

Section 07: The people with a stake

Stakeholder Interest Influence Stake Concern
Priya Nandakumar and Owen Kalu, founders A growing company that stays true to its mission. High High Building the thing they founded the company to replace, or losing the company by refusing.
Lakeshore Ventures, investor Growth and a return on its investment. High Medium Losing a fifth of revenue before the next funding round.
Engineers Building software they believe in. Medium High Being asked to build surveillance; some say they will leave rather than do it.
The three logistics clients Productivity data by worker, by hour. High Medium Margins that depend on productivity, and a supplier that will not provide the data.
Workers who use the app Control over their shifts, and fair treatment. Low High Being watched, scored and ranked through an app they rely on.
Care and retail clients Good scheduling that their staff are happy to use. Medium Medium A supplier whose reputation with workers changes.
Unions and regulators Lawful, transparent and proportionate monitoring at work. Medium Low Monitoring that workers do not know about or cannot challenge.

Section 08: Strategic alternatives

A. Build Pulse as asked

Build productivity scores, idle-time alerts, location tracking and rankings for each worker.

Requires
C$1,800k of development and engineers willing to build it.
Upside
Keeps the three clients and adds the most profit, C$241k a year from 2027.
Risks
Contradicts the mission; some engineers may leave; workers may turn against the app; more care and retail clients may leave than the 3,000 workers forecast; legal exposure if the data is misused.

Evidence: Exhibit 3, Exhibit 5, Exhibit 6, Exhibit 8

B. Build Pulse Lite

Offer team-level productivity data with no rankings, and individual data only for workers who choose to share it.

Requires
C$1,500k of development, and clients who accept data without rankings.
Upside
Keeps two of the three clients, sells to care and retail clients too, and fits the mission.
Risks
Bayline leaves, so the gain is smaller, C$93k a year; the other two may want more later.

Evidence: Exhibit 4, Exhibit 5, Exhibit 6

C. Open an interface for clients

Let clients send Larkspur's data to analytics tools of their own choosing; build no monitoring.

Requires
C$600k of development and an interface that is secure.
Upside
Keeps the three clients at little cost, and Larkspur does not design the monitoring itself.
Risks
The monitoring happens anyway, on Larkspur's data; the gain is small, C$22k a year; the legal adviser's warning still applies.

Evidence: Exhibit 4, Exhibit 6

D. Decline, and win more care and retail clients

Refuse the request, let the three clients go, and spend more on winning clients in care and retail.

Requires
An extra C$300k a year of sales and marketing, and the investor's patience.
Upside
Keeps the mission, the engineers and the trust of workers intact.
Risks
Revenue falls by C$864k and profit by C$948k a year, a loss of C$764k in 2027, a year before the funding round.

Evidence: Exhibit 2, Exhibit 6, Exhibit 9

Section 09: Questions for analysis

Original practice questions written for this case, not IB examination questions. The marks are each question's own allocation, set to the size of the task. Answer first, then open the guidance.

Source: Original practice question
Show content for

Showing everything. HL-only material is marked HL.

  1. Define the term digital Taylorism.

    Define AO1 5.9 Management information systems HL

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    One mark for the use of digital technology to monitor and measure workers' tasks in detail, and one for the link to scientific management: work broken into measured units, with workers controlled through targets and data. Pulse's scores, alerts and rankings are an example.

    [2 marks]

  2. Using Exhibit 1, calculate Larkspur's gross profit margin and profit margin for 2025.

    Calculate AO4 Paper 2 style 3.5 Profitability and liquidity ratio analysis Exhibit 1

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    Gross profit margin = 3,564 ÷ 4,752 × 100 = 75.0%. Profit margin = 184 ÷ 4,752 × 100 = 3.9%.

    Two marks for each: one for the method and one for the answer. The profit margin uses profit before interest and tax, as the guide's formula does. The gap between the two margins shows that most of Larkspur's costs are expenses, mainly engineers, rather than the cost of serving each worker.

    [4 marks]

  3. Using Exhibit 2, calculate the percentage of Larkspur's 2025 revenue that came from the three largest logistics clients.

    Calculate AO4 3.3 Costs and revenues Exhibit 2

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    The three clients have 25,000 workers at C$3.60 a month: 25,000 × 3.60 × 12 = C$1,080k. Share = 1,080 ÷ 4,752 × 100 = 22.7%.

    One mark for the clients' revenue and one for the percentage.

    [2 marks]

  4. Explain two sources of conflict between Larkspur's stakeholders over the Pulse request.

    Explain AO2 1.4 Stakeholders Exhibit 5, Exhibit 7, Exhibit 8, Exhibit 9

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    Two conflicts, each between named stakeholders and explained with the case, for two marks each. Strong examples: the logistics clients want worker-level data, while the workers who use the app do not want to be ranked; the investor wants revenue growth, while the engineers will not build Pulse; the founders' mission conflicts with the clients' demand. Naming stakeholders without the reason for the conflict earns one mark each at most.

    [4 marks]

  5. Using a motivation theory, explain how Pulse could affect the motivation of the warehouse workers it monitors.

    Explain AO2 2.4 Motivation and demotivation Exhibit 3, Exhibit 5

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    Two marks for the theory and two for applying it. Taylor's scientific management would predict that measured targets raise output; Herzberg would warn that close supervision is a hygiene factor that causes dissatisfaction when it is felt as control; at HL, self-determination theory predicts that taking away autonomy weakens the motivation that comes from within. The survey's 46%, who would consider leaving a job that tracked them, is strong evidence.

    [4 marks]

  6. Discuss the benefits, risks and ethical implications of Pulse for Larkspur and its stakeholders.

    Discuss AO3 5.9 Management information systems; 1.3 Business objectives Exhibit 3, Exhibit 5, Exhibit 6, Exhibit 8 HL

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    A strong answer covers each part of the question with the case.

    • Benefits: C$241k a year of extra profit, three clients kept, and data the clients say they need.
    • Risks: engineers leaving, workers turning against the app, more care and retail clients leaving than forecast, and shared responsibility if the data is misused.
    • Ethical implications: monitoring that workers did not choose, decisions about people made by scores, possible bias in what is measured, and a company acting against its own mission.
    • A judgement that weighs them, for example that the ethical cost falls mainly on the people with least power to object.

    Answers that list points without weighing them stay in the middle of the range.

    [10 marks]

  7. Evaluate the option of declining the request and refocusing on care and retail clients.

    Evaluate AO3 1.3 Business objectives; 1.4 Stakeholders; 3.5 Profitability and liquidity ratio analysis Exhibit 1, Exhibit 6, Exhibit 9

    Guidance for a strong answer
    Source: Assessment explanation The platform's guidance, not an official markscheme.

    A strong answer does four things.

    • It quantifies the cost: revenue down C$864k and profit down C$948k a year, a loss of C$764k in 2027.
    • It weighs what declining protects: the mission, the engineers, the workers' trust, the reputation that wins care and retail clients.
    • It considers timing and the investor: a loss a year before the funding round, against a growth target of 20%.
    • It reaches a judgement, perhaps that declining is principled but avoidable if Pulse Lite keeps most of the revenue.

    Answers that treat the decision as purely financial or purely ethical stay in the middle of the range.

    [10 marks]

HL-only questions are hidden in the SL view. Choose HL above to see them.

Section 10: The decision exercise

Decide before you read the debrief. The exercise is the platform's own; it is not an IB task.

Your role

You are Priya Nandakumar, Larkspur's chief executive and co-founder.

The board meets in June: Owen, you and Hannah Osei from Lakeshore Ventures. The clients want an answer within the month, the engineers are waiting to hear whether they will be asked to build Pulse, and the investor wants growth. You wrote the mission on the wall yourself.

The task
Choose one response and defend it to the board, with its money, its people and its principles.
What to produce
A decision memo of about 250 words, giving the response, its effect on profit against the alternatives, its ethical justification and the safeguards you would set.
Time
40 minutes

Choose one of the strategic alternatives and write down your reasons before you open the debrief below. With JavaScript on, this page keeps your decision on this device and shows how the evidence judges the alternative you chose.

Make your decision

Which alternative do you choose?
The alternative you choose, the evidence for it, its main risk, and what would change your mind, at the length the task asks for. 0 words

Section 11: The debrief

The worked analysis, then every alternative judged against the evidence. One defensible reading, not the only one: try your own first.

Read the debrief
Source: Arjun Agrawal framework

Larkspur's dilemma is real because both sides of it are. The clients' revenue matters a year before a funding round, and the mission, the engineers and the workers' trust are the company's advantage, not decoration. The strongest answers compare each response with the loss of the clients, notice how little extra full monitoring adds over Pulse Lite, and treat the ethics as part of the business case rather than an afterthought.

The worked analysis

DataWhat the exhibits say

  • The three logistics clients pay C$1,080k a year, 22.7% of revenue (Exhibit 2).
  • Larkspur's gross profit margin is 75.0% and its profit margin 3.9%, so losing those clients would turn its first profit into a loss (Exhibit 1).
  • From 2027, Pulse would add C$241k a year to profit, Pulse Lite C$93k and the interface C$22k, while declining would cost C$948k (Exhibit 6).
  • 81% of surveyed workers would be uncomfortable being ranked, and 34 of the 58 engineers signed the letter against Pulse (Exhibits 5 and 8).

InsightWhat it means

  • The real comparison is not with today but with losing the clients. Against declining, Pulse is worth C$1,189k a year and Pulse Lite C$1,041k, so the extra profit from full monitoring is only C$148k a year.
  • Pulse is digital Taylorism. It measures workers minute by minute and ranks them, which is the opposite of the autonomy that Larkspur's mission promises and that workers say they value.
  • The interface looks like a way to keep the clients without building the monitoring, but the monitoring would still run on Larkspur's data, and the legal adviser's warning applies to it too.

OptionsWhat each choice does

  • Building Pulse earns most and puts the company's mission, engineers and reputation with workers at risk for C$148k a year more than Pulse Lite.
  • Pulse Lite keeps most of the revenue, fits the mission and sells beyond logistics, at the cost of Bayline.
  • The interface keeps the clients cheaply and leaves Larkspur's data doing the watching.
  • Declining keeps everything except the money, and the money matters before a funding round.

RecommendationOne defensible answer

One defensible recommendation is Pulse Lite. Offer the logistics clients team and site data, no rankings, and individual data only where a worker chooses to share it, and sell the same product to care and retail clients as evidence that productivity and fairness can go together. Publish the rules on what Larkspur's data may be used for. Accept that Bayline may leave, and tell the investor why the company's advantage, the trust of the people who use it, is worth more than the difference.

CaveatWhen it would change

This depends on Northgate and Halvorsen accepting data without rankings, which the product team expects but has not confirmed. If all three would leave rather than accept Pulse Lite, the choice becomes a stark one between the mission and survival before the funding round, and the founders would have to decide whether a smaller, consistent Larkspur, funded for longer, is a company the investor will back.

Every alternative, judged

  1. A. Build Pulse as asked

    Defensible if its conditions hold

    The most profitable response, C$241k a year, and it keeps all three clients. But compared with Pulse Lite it adds only C$148k a year, and it buys that with digital Taylorism the mission rejects, an engineering team that may lose members, workers 81% of whom do not want to be ranked, and legal exposure. Defensible only if the board believes the non-financial risks are small, which the evidence does not suggest.

  2. B. Build Pulse Lite

    Best supported by the evidence

    It keeps two of the three clients and most of the revenue, gives workers control over their own data, fits the mission and can be sold to care and retail clients as well. Its gain is smaller, C$93k a year, because Bayline leaves; against declining it is still worth C$1,041k a year. It depends on Northgate and Halvorsen accepting data without rankings.

  3. C. Open an interface for clients

    Weakly supported by the evidence

    It keeps the clients at little cost and lets Larkspur say it built no monitoring. But the monitoring runs on Larkspur's data all the same, the gain is only C$22k a year, and the legal adviser's warning still applies. It avoids the decision rather than making it.

  4. D. Decline, and win more care and retail clients

    Weakly supported by the evidence

    The cleanest response in principle and the costliest in practice: revenue down C$864k, profit down C$948k a year and a loss of C$764k in 2027, a year before a funding round that needs growth of 20%. Principled, but unnecessary if Pulse Lite can keep most of the revenue.

The trade-offs

  • Revenue against mission. The most profitable response contradicts what the company exists to do.
  • The clients against the users. The people who pay and the people who use the app want opposite things.
  • Now against later. Declining protects the long-term advantage and endangers the short-term funding.
  • Control against responsibility. The interface hands control to clients without handing over the responsibility.

The evidence that matters most

  • The three clients' share of revenue, 22.7% (Exhibit 2).
  • The small difference between Pulse and Pulse Lite, against the large cost of declining (Exhibit 6).
  • The workers' and engineers' views (Exhibits 5 and 8).
  • The legal adviser's warning, in the constraints.

What would change the recommendation

  • All three clients refuse Pulse Lite. The choice becomes Pulse or declining, and the funding round decides it.
  • Care and retail clients ask for worker-level monitoring too. The mission's commercial value is less than it seemed.
  • The investor agrees to fund a smaller, slower Larkspur. Declining becomes affordable.

Common errors

  • Comparing each response with 2025 rather than with losing the clients.
  • Treating ethics as a final paragraph rather than a cost and a risk in the decision.
  • Assuming that the interface removes Larkspur's responsibility for how its data is used.
  • Defining digital Taylorism without showing how Pulse fits it.

Section 12: Theories and frameworks

The theories and tools this case calls for, each with how it applies here and where the platform teaches it.

HL-only frameworks are hidden in the SL view.

Section 13: For seminar discussion

Prompts for undergraduate and MBA seminars. They go beyond the IB course and are labelled as enrichment.

Source: Enrichment
  1. Is a supplier responsible for how its customers use its product? Compare building Pulse with opening the interface, and say where you would draw the line.
  2. Larkspur's investor backed it because workers love the app. How should a founder value an advantage that does not appear in the accounts when it conflicts with revenue that does?
  3. Design the rules you would publish for Pulse Lite: what data, seen by whom, for what decisions, and how a worker could challenge a decision made from it.