FICTION · An Unfair Advantage Intervention
Fiction. Real history. Sourced below.
One of the Circuit City workers fired in March 2007 had seven years of service and earned $11.59 an hour. The federal average for a retail salesperson at the time was $11.14. The company removed an experienced, product-knowledgeable employee to save about forty-five cents an hour — which is only a rational trade if you believe his replacement will sell exactly as much. Nobody believed that. Nobody had measured it either, and that is the whole case.
I’m Mark Mitchell, and I’ve run the finance function through cost reductions, so I want to be careful not to make this a story about cruel executives and sympathetic workers, satisfying as that is. It’s a story about a measurement system. Circuit City could see exactly what each associate cost and had no way at all to see what each associate produced — and an organization will always act on the number it can see.
Full fiction disclosure and sourcing note
This is a work of fiction. Mark Mitchell is a fictional character from The Unfair Advantage by Todd Hagopian, and Cartwell is the book’s fictional company. Mark analyzes the real, documented history of Circuit City from the present day; he never advised the company, never met or spoke with Philip Schoonover or any other real person, and no dialogue is invented for them. Statements by fired employees are quoted from contemporaneous news reporting. The historical facts are accurate and sourced as of 2026; the framework analysis is invented.
- Forty-Five Cents an Hour
- Revenue Responsibility Engineering: The Missing Column
- The 2003 Decision That Made 2007 Inevitable
- What Best Buy Was Doing With the Same Years
- The Execution Was Its Own Decision
- What Actually Happened
- From the Operator’s Chair
- The Monday Morning Playbook
- The Contrarian Reappraisal
- The Circuit City Timeline
- By the Numbers
- Frequently Asked Questions
- About the Stagnation Assassin
Forty-Five Cents an Hour
On March 28, 2007, Circuit City fired about 3,400 store employees — roughly 8% of its workforce — and said plainly that the dismissals had nothing to do with performance. They were paid too much.
The mechanism was a pay cap set for each department. Anyone earning more than 51 cents above their department’s cap was terminated. One worker told the Washington Post that the cap for his full-time computer department role was $15.50 an hour and he earned $18.72. Seven years of service. Wages across those fired ran roughly $10 to $20 an hour.
Here is the number that should end any argument about whether this was analysis. Another fired worker with seven years of service earned $11.59 an hour, against a Bureau of Labor Statistics average for retail salespeople of $11.14. Circuit City removed him to capture about forty-five cents an hour.
That trade is rational under exactly one assumption: that his replacement — the company said openly that new applicants needed no sales experience — would generate identical revenue. Nobody at Circuit City believed that. The point is that nobody had any way to check, and that is a design flaw in the measurement system, not a character flaw in the executives.
Revenue Responsibility Engineering: The Missing Column
Revenue Responsibility Engineering exists to correct one specific and extremely common category error: treating people who generate revenue as a cost line, because the cost is measurable and the revenue contribution isn’t. Its four phases are Visibility, Accountability, Incentive Alignment, and Cultural Embedding, and the order matters — you cannot hold anyone accountable for a contribution you cannot see.
Circuit City failed at Phase 1, and everything else followed mechanically.
Consider what its system could actually observe about a sales associate in early 2007. It could see an hourly wage. It could see a department pay cap. It could compute the difference to the cent — 51 cents, $3.22, whatever it happened to be. What it could not see, anywhere, was how much that person sold, what attachment rate they achieved, how many difficult high-margin decisions they closed, or how many customers returned because of them.
An organization acts on the numbers it can see. Give management a payroll system with a precise cost figure and no revenue figure, and the only available action is a cut — not because anyone is callous, but because a cut is the only operation the data supports. The savings projection was rigorous: $110 million in that fiscal year and $140 million annually thereafter, from the firings together with several other moves. Every dollar of that was modeled. The other column wasn’t modeled at all, because it didn’t exist as data.
And the company half-knew it. Circuit City itself warned that sales might be volatile in the first half of the year while new salespeople learned their jobs. That is an acknowledgment that revenue would be affected, offered without a number attached — a qualitative caveat next to a quantified saving. In any decision, the quantified side wins.
The 2003 Decision That Made 2007 Inevitable
Here is the causal chain the standard telling misses entirely, and it’s the most useful thing in the case.
Circuit City eliminated its commissioned salesforce in 2003. That decision is usually filed alongside 2007 as another instance of the same error, but it is better understood as the cause of it. Commission is a measurement instrument. Under commission, every associate’s revenue contribution is visible, continuously, in the payroll system itself — the two columns sit side by side, and a high earner is visibly a high producer.
Remove commission and the revenue column disappears while the cost column remains. From 2003 onward, the only quantified fact Circuit City held about any associate was what they cost. A tenured, expert, high-selling associate and a tenured, mediocre, expensive one became indistinguishable in the data — both simply appeared as people whose wages had crept above a benchmark.
Four years later, management looked at that data and did the only thing the data permitted. The 2007 cut was not a fresh error. It was the 2003 measurement change coming due.
The second-order effect is worse and it’s Phase 3 running backwards. Once associates understood that wages above a cap were dangerous, the incentive pointed away from excellence. Selling exceptionally well over years produced raises, and raises produced exposure. An incentive system that makes outperformance hazardous doesn’t merely fail to motivate — it actively selects against the behavior you need, and it does so quietly, for years, before anyone notices.
What Best Buy Was Doing With the Same Years
Magnificent Obsessions has two pillars, and the one that matters here is Competitor Obsession — deconstructing a rival’s business model rather than tracking its surface behavior. Circuit City watched Best Buy and read it as a discounter to be matched on price. That reading was wrong at the level of model.
In the same years, Best Buy acquired Geek Squad and built a genuine service-and-support business, reorganized stores around customer segments, invested in locations and its website, and managed high-risk inventory. It was building revenue-generating capability into the store experience. Circuit City read the price tags.
The compensation comparison sharpens it. Circuit City’s chief executive was paid $8.52 million in fiscal 2006, against $3.85 million for Best Buy’s chief executive, in a year Circuit City lost $11.8 million. Two companies, same industry, same pressures, and the one paying its leadership less than half as much was the one investing in capability.
And the cost-center reflex wasn’t confined to the sales floor. The same March 2007 announcement outsourced Circuit City’s IT infrastructure to IBM, affecting 130 headquarters workers. Two functions, one instinct: if you can measure what it costs and not what it produces, it’s a cost.
The Execution Was Its Own Decision
Everything above concerns the analysis. The execution was a separate decision, made separately, and it compounded the damage well beyond the capability loss.
Employees were notified in morning meetings and escorted from the stores the same day. They received four weeks of severance and were told they could reapply for their old jobs after a ten-week waiting period, at reduced wages, with no guarantee the job would exist. Among them were people with two decades of service.
One worker with seven years at the company told the Post he had dedicated seven years and that loyalty got him nothing. That sentence travelled, and it was heard by every remaining employee and a great many customers.
From a measurement standpoint this is the part that never appears in any model, and it is not unmeasurable in principle — it’s just not on the sheet. What the remaining associates learned that morning was the actual compensation policy: tenure is a liability and expertise is priced at the market rate for inexperience. You cannot then ask those same people to deliver consultative service, because you have just told them precisely what it’s worth.
What Actually Happened
On March 28, 2007, Circuit City fired approximately 3,400 store employees, about 8% of its workforce, stating the dismissals had nothing to do with performance and that the workers earned above the market-based range for their roles. Anyone more than 51 cents above their department’s pay cap was terminated. Workers were notified in morning meetings, escorted out the same day, given four weeks’ severance, and invited to reapply after ten weeks at lower wages; the company said new applicants needed no sales experience.
The firings, together with the outsourcing of IT infrastructure to IBM, the closure of seven U.S. stores, a distribution center, and 62 Canadian stores, were projected to reduce expenses by $110 million in fiscal 2008 and $140 million annually from fiscal 2009. Circuit City had lost $11.8 million in the fiscal year ending February 28, 2007. Its shares rose on the announcement.
The layoffs came after a decade of compounding errors: the failed DIVX disc format, the 2000 exit from an appliance business representing more than a billion dollars in annual revenue just before the mid-2000s housing boom, the 2003 elimination of the commissioned salesforce, aging stores in weaker locations than Best Buy’s, and substantial share repurchases that left little cushion.
The 2008 financial crisis froze vendor credit, leaving Circuit City unable to stock shelves for the holidays. It filed for Chapter 11 on November 10, 2008. Liquidation was approved in January 2009, and all 567 remaining stores closed by March 8, 2009, eliminating more than 35,000 jobs. Systemax purchased the Circuit City brand, trademarks, and domain for $14 million in May 2009.
From the Operator’s Chair
I’m Todd Hagopian, and I’ve had to cut costs in losing businesses across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel. This is the case I revisit before any hard people decision, because the trap in it is one I could fall into. Three things.
First, before you cut a group of people, ask what your system can actually see about them. If the answer is cost and only cost, you are not making a decision — you are executing the only operation your data permits, and you will find out what you destroyed afterward. Circuit City could compute a pay-cap variance to the cent and had no figure at all for revenue contribution. Get the second column built before you need it, because you will never build it in the middle of a cost crisis.
Second, understand that removing a measurement is a decision with a four-year fuse. This is the part I’d want every finance leader to sit with. Eliminating commissions in 2003 looked like a compensation change; it was actually the deletion of the only instrument that made individual revenue contribution visible. Once that was gone, the 2007 layoff was close to inevitable, because expensive-and-excellent and expensive-and-mediocre had become the same row in the same report. When someone proposes simplifying a measurement, ask what decision becomes impossible to make well once it’s gone.
Third, price the message you send, because it is a real cost even though it never appears on the sheet. Escorting twenty-year employees out in the morning and inviting them to reapply at lower pay taught every remaining associate the company’s actual view of tenure and expertise. You then cannot ask those people for consultative selling — you’ve just quoted them the price of it. You can reduce costs and keep your honor. Doing so is slower and more expensive in the quarter, and it is the only version that leaves you with a company worth having afterward.
The Monday Morning Playbook
- Ask what your system can see before you cut. If you have a precise cost figure and no revenue-contribution figure for a group of people, you don’t have an analysis — you have the only operation your data allows.
- Never delete a measurement without naming what it made visible. Circuit City’s 2003 commission change removed the instrument that showed individual revenue contribution, and the 2007 cut followed directly from its absence.
- Check whether your pay structure punishes excellence. If crossing a pay cap is dangerous, your best people will learn to stop outperforming — quietly, years before anyone notices.
- Deconstruct your competitor’s model, not their prices. Circuit City read Best Buy as a discounter while Best Buy was building a service business. Surface behavior is not a business model.
- Treat the execution as a separate decision with its own cost. How you cut teaches everyone who remains what you actually value, and that lesson outlives the savings.
The Contrarian Reappraisal
The standard reading — greedy executives fired their best people and the company died — is right in spirit and too simple to use.
The economics were smaller and worse than the story suggests. One fired worker with seven years of service earned $11.59 an hour against a $11.14 national average: a forty-five cent margin. The projected $110 million came from the firings plus store closures, a distribution center, Canadian closures, and an IBM outsourcing — not from the layoffs alone. This wasn’t a bold if brutal cost play. It was a small saving purchased with a large capability.
The 2007 decision was made in 2003. Eliminating the commissioned salesforce removed the only mechanism that made individual revenue contribution visible. From then on, the sole quantified fact about any associate was cost, and the expensive expert and the expensive mediocrity became the same row. Management in 2007 acted on the only data it had.
The layoffs didn’t kill the company; they were the most legible symptom. DIVX, the exit from a billion-dollar-plus appliance business before the housing boom, weaker store locations, share repurchases that left no cushion, and a 2008 credit freeze that emptied the shelves all did their work. But the layoffs are the one everyone remembers, because they’re the one where the missing column is visible to the naked eye.
The Circuit City Timeline
- 1999 — Circuit City writes off the failed DIVX disc format, burning capital and credibility as Best Buy begins building services.
- 2000 — it exits major appliances, more than a billion dollars in annual revenue, just before the mid-2000s housing and appliance boom.
- 2002–2003 — Best Buy acquires Geek Squad and builds a service business; Circuit City eliminates its commissioned salesforce, deleting the instrument that made individual revenue contribution visible.
- Fiscal 2006 — Circuit City’s CEO is paid $8.52 million against $3.85 million for Best Buy’s; Circuit City loses $11.8 million in the year to February 2007.
- March 28, 2007 — about 3,400 associates more than 51 cents above department pay caps are fired in morning meetings and escorted out, invited to reapply in ten weeks at lower pay.
- 2007 — the same announcement outsources IT infrastructure to IBM and closes seven U.S. stores, a distribution center, and 62 Canadian stores, projecting $110 million in savings.
- November 10, 2008 — with vendor credit frozen and shelves unstocked for the holidays, Circuit City files for Chapter 11.
- March 8, 2009 — all 567 remaining stores close, ending more than 35,000 jobs; the brand sells for $14 million in May.
By the Numbers
- ~45 cents an hour — the margin between one fired seven-year employee’s $11.59 wage and the $11.14 national average for retail salespeople.
- 51 cents — how far above a department pay cap an employee had to be to lose their job.
- 3,400 / 8% — associates fired in a single morning, and their share of the total workforce; the company said performance had nothing to do with it.
- $110M / $140M — projected savings for fiscal 2008 and annually thereafter, from the firings together with store closures, a distribution center, Canadian closures, and an IBM outsourcing.
- $8.52M vs $3.85M — Circuit City’s CEO compensation against Best Buy’s in fiscal 2006, in a year Circuit City lost $11.8 million.
- 567 stores, 35,000+ jobs, $14 million — what closed by March 2009, and what the brand later sold for.
Frequently Asked Questions
Why did Circuit City fire its highest-paid employees?
On March 28, 2007 it terminated about 3,400 store employees — roughly 8% of its workforce — who were paid more than 51 cents above the pay cap set for their department, stating the dismissals had nothing to do with performance. They were notified in morning meetings, escorted out the same day, and invited to reapply after ten weeks at lower wages. The company said new applicants needed no sales experience.
How much did Circuit City actually save?
The firings, combined with outsourcing IT infrastructure to IBM and closing seven U.S. stores, a distribution center, and 62 Canadian stores, were projected to cut expenses by $110 million in fiscal 2008 and $140 million annually thereafter. The per-worker margins were remarkably thin: one fired employee with seven years of service earned $11.59 an hour against a national retail average of $11.14 — a difference of about forty-five cents.
Did the 2007 layoffs cause Circuit City’s bankruptcy?
They accelerated and symbolized it rather than causing it. Circuit City had already written off the failed DIVX format, exited an appliance business worth more than a billion dollars in annual revenue just before the housing boom, eliminated its commissioned salesforce in 2003, operated weaker store locations than Best Buy, and spent heavily on share repurchases. The 2008 credit freeze left it unable to stock shelves for the holidays, and it filed for Chapter 11 that November.
How did Best Buy survive when Circuit City didn’t?
By building revenue-generating capability where Circuit City cut it. In the same years Best Buy acquired Geek Squad and built a service business, reorganized stores around customers, and invested in locations and its website. Circuit City read Best Buy as a discounter to be matched on price and missed that it was changing its business model. Best Buy’s CEO was also paid less than half what Circuit City’s received.
What’s the real lesson of the Circuit City layoffs?
That organizations act on the numbers they can see. Circuit City could compute each associate’s cost to the cent and had no measure at all of their revenue contribution — largely because eliminating commissions in 2003 had deleted the instrument that made contribution visible. Four years later the only available action was a cut. Build the revenue column before you need it, because you will not build it during a cost crisis.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B+ in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and JBT Marel, where he serves as VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he turns proprietary frameworks — including Revenue Responsibility Engineering and Magnificent Obsessions — into field-tested tools for operators deciding which costs are actually capabilities. He is the author of The Unfair Advantage and the Stagnation Assassin series, a Forbes contributor, and an award-winning keynote speaker. Connect with Todd on LinkedIn or bring him to your stage via his speaking page.
Forty-five cents an hour, against seven years of product knowledge. That trade only looks sane if you can see one column and not the other — and Circuit City had deleted the other one four years earlier. Build the revenue column before you need it, because you will never build it in a cost crisis.
The discipline of telling costs from capabilities is in The Unfair Advantage.