Commerce Bank · Prep

The Professor's Questions

Service Excellence by Design. The case takes place as a technological breakthrough made it possible for customers to bank without visiting a physical branch. Commerce Bank was one of the fastest-growing banks in the United States by deposit growth, and had launched a pilot called Retailtainment.

Analyze Commerce Bank's service delivery system before Retailtainment. How well did its operating model support its competitive strategy? Which design choices were most critical to the bank's success?
What problem was Retailtainment designed to solve? How successful was it? What, if anything, would you change?
How should competitors have responded to Commerce Bank's market entry?
What lessons from the rise of online banking are most relevant to the age of AI?

My Answers

Q1. Operating model vs strategy

The operating model and the strategy fit tightly — better than almost any retail bank in the case. Commerce does not try to be a nicer version of Citibank. It is a retailer that happens to take deposits. Every major design choice feeds one job: get people to walk in, open a core deposit, and stay.

The strategy in one line: win on service and convenience (the 62%), not on rate (the 3%). Pay about 50 bp less on deposits. Spend the savings on hours, people, and small delights — and give cost wins back to the customer (free ATM and debit, NYC surcharge rebate up to $5/month) instead of dropping them to the bottom line.

The design choices that actually matter, in order:

The store is the product. 7:30 a.m.–8 p.m., weekends, midnight drive-through, the 10-minute rule. Average 40,000 visits a month, 100,000 at the busy ones — more than a McDonald's. Hill is explicit: even if nobody banked on Sunday, Sunday hours would still work, because they say "I am always there."
One prototype, copied. ~$1 million, built from scratch, same white brick, black roof, marble, lollipops, dog biscuits. "We know every screw." Break-even in 12–18 months versus three years for the industry. Consistency cuts customer learning cost and makes training, mystery shopping, and hiring scalable.
A short menu. Four checking products, not a dozen. After a merger, other banks force customers into unfamiliar products. Commerce never creates that mess. Simplicity is what lets a "wack job" hired for warmth still get the transaction right.
Incentives face the same way as the strategy. Deposits and loans both credit the store. No deposit, no loan. The loan officer is the branch manager. WOW shops (14,000 a year) drive raises. Retailer of the Year is scored on core deposit growth, WOW shops, losses, and turnover — not on calls handled. Fleet's healthcare team keeps the deposit credit while the branch does the work; Commerce inverted that, and the politics flipped.
People system, not posters. Commerce University, 41 people. Traditions teaches the handshake and "smiles have teeth." 2,000 interviews for 40 Manhattan jobs. Mystery shoppers ding you for saying "can" instead of "may." Kill-a-stupid-rule pays $50. This is how the cult survives growth.
Credit is conservative on purpose. Loan-to-deposit 45.5% versus the industry's 87%. Non-current loans 0.38% versus 1.44%. They do not earn enough spread to buy credit risk. The deposit franchise is the asset; securities are 42.5% of assets versus 18% for the industry. They are a deposit gatherer that happens to lend, not a lender that happens to have branches.
Do not push people off the expensive channel. Industry: advertise the ATM, sometimes charge $3 to see a teller. Commerce: every channel should be excellent. Online usage still lands at 34%, above Wells Fargo. The cheap channel is a complement when the expensive one is loved, and a rejection when it is used as a dump.

Fit is visible in the P&L. Efficiency ratio 70.3% versus 59.7% — they are more expensive, as promised. Cost of interest-bearing deposits 2.37% versus 5.35%. ROE 15.0% versus 12.4%. Net income doubled from 1998–2001 while the industry grew 20%. Deposit growth in 2001 was 37.5% against the industry's 5.1%. The operating system is not a decoration on a bank. It is the strategy, running.

The critical few: hours + identical stores + four products + store-level deposit credit + the WOW people system + "we do not push you out." Take any one away and the 62% story starts to leak. Retailtainment is not on this list.

Q2. Retailtainment: problem, results, what I would change

Retailtainment was designed to solve a next-move problem, not a waiting-line problem. Competitors had started copying the easy layer of Commerce's offer — evening and weekend hours, "human" ads, a café, TVs in the queue. Pauls says it out loud: "We need the next penny arcade." The bank pays less on deposits on purpose; if the service gap closes, the 50 bp discount becomes a reason to leave. Friday-afternoon experiments were a search for a new, visible wow, plus a bet that local managers know their fans better than Cherry Hill does.

It was not successful, on the evidence in the case. We get no lift in deposits, no cut in perceived wait, no mystery-shop scores. We get: a greeter problem ("too many greeters standing around"), a friend who would rather the greeter get behind the counter, and a hot-dog cart that catches fire, takes the juggler with it, and makes the local news. The one thing Commerce had been religious about — one bank, one brand, every screw the same — is the first thing Retailtainment relaxes.

What I would change: kill it as a local variety show. Keep the search for the next penny arcade, but run it the way they already know how to run a store.

Standardize delight; do not freelance it. Penny Arcade, umbrellas, the 10-minute rule, check images the next day, a phone on the ATM — these are designed once, copied everywhere, and they solve a real job (coins, rain, "are you open?", "did my check clear?"). A guitarist on Friday solves none of those jobs and cannot be mystery-shopped.
If you must prototype, prototype like a retailer, not like a talent show. One idea, designed by marketing and operations together, piloted in a handful of stores, with a script, a safety check, and a kill rule. The hot-dog cart had marketing approval and still burned. Approval without a playbook is not control.
Put the greeter back on the value stream. The friend's complaint is the real operations issue: peak-load capacity. 40,000–100,000 visits a month means the constraint is tellers, not atmosphere. A Friday juggler while the line is stuck is a strategy–ops contradiction. Flex staff into the queue before you flex staff into costume.
Measure it or it is theater. Add Retailtainment to WOW-shop items only if there is a single expected behavior. If there isn't, it does not belong in a system whose raises already hang on 14,000 shops.
Do not spend the "next penny arcade" budget on copying WaMu's children's corner. The next penny arcade should be another job-to-be-done that rivals refuse (they already refused coins). It should not be a louder lobby.

The honest sentence for Jacovelli: Retailtainment is a symptom that the leadership team is anxious about imitation. The cure is to go deeper on the uncopyable system — hiring 50-to-1, four products, store-level P&L, Sunday hours — not to go louder on the copyable theater.

Q3. How competitors should have responded

They should not have copied the lobby. That is what they did — WaMu's un-bank, ING's espresso, Bank of America's TVs — and it is the weakest possible reply, because it copies the one layer Commerce can replace next Friday.

Three better replies, depending on who you are:

If you are a money-center bank (Citi, BofA, Fleet): do not get on Commerce's field. You cannot retrofit a cult. Hill is right about the direction even if he overstates the history: "You can only build a delivery model like this from scratch." Mergers already wrecked your service (34% of leavers cite fees, service, errors; Robbins walks across the street at 8:15 a.m.). Adding a juggler on top of a merger integration is how you become a punchline. Pick a different excellence and say it: best rate, best digital, best for companies, best for the top 5% of balances. Then design down to that — including the things you will be bad at. Charge for tellers if you mean it. Close on Sunday if you mean it. Stop running ads that claim Nordstrom while the operating system still pays on call volume.
If you insist on fighting for the 62%: copy the trade-offs, not the mascot. Match hours. Cut the product list to a handful. Give the branch the deposit credit. Build a university. Shop yourself 14,000 times. Interview 50 people for every teller. Pay less on deposits and publish it as the price of the experience. Accept a 70% efficiency ratio. Stop merging for cost saves in the same metro where Commerce is opening. The $5,000 Commerce pays its staff when you close a branch is a clue: your exits are their marketing. None of this is a campaign. It is a five-year rebuild. Most boards will refuse it. That refusal is the response — concede the service segment and harvest the 3% who want rate.
Use the two leakages Commerce is already fishing. 34% of bank leavers move house; 34% leave because of fees and bad service, often right after a merger (Exhibit 8). A competent incumbent would: keep surplus branches for 18 months after a merger instead of "rationalizing" them into Commerce's lap; waive the surprise fees that create the 34%; and staff the remaining stores on Saturday. You do not need Mr. C to stop a geographic and a post-merger leak.
Do not try to win New York by out-promoting $500,000 of hot dogs. Commerce broke even in half the time at 4× cost because the model was already loved. A copycat without the model is just a more expensive me-too, which is what Hill founded the bank not to be.

The response that looks aggressive and is actually weak: match Sunday hours and install a coffee machine. The response that looks passive and is actually hard: choose a different axis of excellence, and stop pretending you compete on service when your economics require pushing people to the VRU.

Q4. Lessons from online banking for the age of AI

The online-banking story in this case is not "digital won." It is: banks used a new technology to push customers where the bank wanted them to go, and Commerce used the same technology to go where customers already were — and got more digital use, not less.

The industry logic in 2002: teller is expensive, so advertise the ATM, the VRU, the website; pay people to move; sometimes fine them $3 for coming in. Commerce's logic: you cannot name a retailer that survived by driving people where they do not want to go. Offer the best of every channel. Result: 34% online usage, above Wells Fargo, and 40,000–100,000 store visits a month.

That is the AI lesson. Four specifics:

Do not confuse a cheaper channel with a better offer. AI will make self-service, chatbots, and "digital-only RMs" look free at the margin, the way the website looked in 2002. If you push clients there to save on the human, you are Fleet charging $3 for a teller. The clients who can leave, will. Commerce's 34% says the opposite: people adopt the new channel when the old one still loves them.
Use the new tech to make the expensive channel excellent, not empty. Commerce put a phone on the ATM so you could reach a human; showed the cheque image the next day; let you opt out of the VRU without a hunt. That is High Tech in service of High Touch. The AI analogue: prep the meeting, remember the family, draft the memo, sit in the room as a second memory — do not replace the room.
Gimmicks arrive first; systems arrive late. Retailtainment is the 2002 version of "we should have an AI demo on Friday." Penny Arcade is the 2002 version of a real workflow (count the coins, print a receipt, no fee). In the age of AI, refuse the juggler. Fund the coin machine: one job, same in every "store," mystery-shoppable.
You still cannot be great at everything. Frei's class title is Service Excellence by Design. Design means trade-offs. Commerce is great at convenience and average-to-poor at rate and at credit risk. AI does not lift that constraint. If you use AI to try to be the cheapest and the warmest and the most complete, you will get a chatbot that apologizes in four products' jargon. Pick the 62% you serve, and let AI deepen that, including the parts you will stay bad at.
Consistency scales; personality-as-exception does not. 185 stores, one greeting, 14,000 shops. AI makes it tempting to let every RM (or every model) improvise. That is Retailtainment at software speed. The lesson is: write the playbook first, then let the machine copy it. Kill the stupid rule remains a human job.

One sentence for class: the banks that "won" at online were not the ones that moved customers online; they were the ones customers still wanted to visit — and then followed online. AI will rhyme.

Case Summary

Halloween 2002, Cherry Hill, New Jersey. A pumpkin, a Dalmatian and a masked avenger run past Deborah Jacovelli's office door. This is Commerce University — Commerce Bank's full-time training center. The giant letter C outside has been draped in a cape. Jacovelli is the dean here. She is used to this culture that looks nothing like a bank's, but something new this year makes her uneasy: headquarters has just launched a pilot called Retailtainment, which lets branch managers experiment freely on Friday afternoons with ways to entertain waiting customers — free hot dogs, guitarists, jugglers, someone in a giant C costume on roller skates. Some stores have already had incidents: a hot dog cart caught fire, a juggler got burned, and it made the local news.

Vernon W. Hill II opened this community bank in southern New Jersey in 1973 with $1.5 million. His judgment: the world did not need another me-too bank. He had no capital and no brand, so he had to find a way to be different. Over the next thirty years Commerce made no acquisitions, opening its own stores into Pennsylvania, Delaware and New York. Hill called his branches stores, and benchmarked against Starbucks and Home Depot, not the bank next door. He asked: why will Starbucks get you to pay $6 for a cup of coffee? Because of the retail experience. Since 1990 the stock had risen 20-fold. CFO Douglas Pauls's words are on the cover: "The hardest thing about becoming a big bank is not becoming a big bank." — the hardest part isn't growing up, it's growing up without turning into a big bank.

Around 2001, the industry was solving a different problem. Deposits and loans were both commodities. From 1998 to 2001, industry deposits grew about 19% cumulatively, and nearly 90% of deposits were lent out; the outsized growth of the big banks came almost entirely from acquisitions. Two trends: pushing cross-selling up from 1.5–2.5 products per customer; and leaning on fees — non-interest income +27%, interest income only +11%. On service, banks began tiering customers by profitability; satisfaction was chronically low, with 79% "very satisfied" at credit unions versus only 53% at retail banks. Attrition at the worst could reach a third of customers a year, and even the best lost 15%. Of those who left, 34% did so because of fee surprises, poor service or errors, 34% because they moved outside the branch network, and 15% because somewhere else was more convenient or open longer. On channels, the total number of branches kept rising even as the number of banks fell; ATMs went from cost centers to fee-generating profit centers; almost every bank had a 24/7 call center and online banking. Because electronic channels had low marginal costs, most banks actively pushed customers from the teller window to self-service — through advertising, rewards, and occasionally penalties: $3 to see a teller. Front-line hiring targeted people who could do repetitive work and accept low wages; performance was measured by call volume and transaction counts.

Commerce turned this whole logic on its head. Stores were open 7:30–20:00 on weekdays and on weekends too; busy drive-through lanes could stay open until midnight. There was also a 10-minute rule: open 10 minutes early, close 10 minutes late. There were only 4 kinds of checking accounts, and opening one came with a gift. In Hill's words: other banks, because branches are expensive, try to push people out and online; "You can't find a successful retailer that survives by pushing people to where they don't want to go." Their model was to be the best in every channel, and to assume you'd use all of them. The result was not only the fastest deposit growth in America, but an online usage rate of 34% — higher than Wells Fargo's. An average store had 40,000 visits a month and a busy one 100,000 — McDonald's gets about 25,000. Hill said: I don't have to resell you every day; once you've opened an account, your balance is making me money. Big banks saw customers as a cost. He didn't.

Growth was organic, and New York was organic on steroids. Others said New York was the most crowded banking market in America; Hill said it was the most underserved — there was $500 billion in deposits there. Each store got a $500,000 promotion, five times the usual: direct mail, subway ads, phone booths, plus street vendors handing out 10,000 hot dogs wrapped in Commerce napkins. Costs were four times those of other regions, but the payback period was cut in half. Hill was adamant about not growing through acquisitions: "No power retailer in this country was built by acquisition. This delivery model can only be built from scratch. Acquisitions are at heart about cutting costs; culture and brand get diluted. It's a formula for failure. Every big bank merger in this country has failed." "It's easier to build a bank than to fix one." It took the company 18 years to reach $1 billion in deposits; now it could exceed that in a single quarter. The goal was $100 billion in deposits and 1,000 stores, from Washington to Boston. Net income doubled from 1998 to 2001, versus +20% for the industry. When Pauls joined in 1994 he thought the projections were aggressive; later the whole plan was left in the dust.

What they sold wasn't loans — it was core deposits. Hill: a bank's value isn't in its loan book, it's in deposits that didn't come for the interest rate. They were typically the lowest payer in the market, often 50 basis points lower. The consumer business contributed more than half of deposits, versus about a third for the industry. In the '80s and '90s the big banks gave up on gathering retail deposits, funding themselves in wholesale markets while cutting back their retail networks. Pauls said Vernon wasn't afraid of competition — rivals were fighting an air war, while they fought a ground war in the stores. Longer hours are of course more expensive, and the expense ratio would look ugly. Hill's reversal: "A low expense ratio is a negative, not a positive. People with low expense ratios are draining capital from their business every day." Asked why they opened a new account, 3% of customers said they wanted the highest rate, and 62% said service and convenience. Rivals fought over the 3%; Commerce fought for the 62%. Money saved didn't go to the income statement — it went back to customers. Switching ATM contracts saved several million, which they used to abolish card fees and transaction fees company-wide; in New York, when someone else's machine charged you $1.50, they reimbursed you (up to $5 a month). Hill said this is how a power retailer operates: use your competitive advantage to make yourself stronger, not to fatten your margins.

Loans were a supporting act, and designed to pull in yet more deposits. Loans were credited to the store serving that customer, and so were deposits. Dennis Falese contrasted this with Fleet: deposits brought in by the medical lending group were credited to the group, while the branches still had to service them and got no credit, so branch managers hated the lending group; at Commerce it was the reverse — branch managers loved the lending people. The loan-to-deposit ratio was well below the industry's. Falese: "The spread we earn doesn't cover the credit risk. Whether the customer can repay comes first. Some of our best loans are the ones we didn't make." The policy was no deposit, no loan. Loan delivery was pushed down to the stores; at some point you had to come into the store to sign, and the loan officer was the store manager — who used that moment to persuade you to bring your deposits over as well.

Stores were built as replicable retail machines. Floor-to-ceiling windows, plenty of parking, built from scratch for about $1 million — "we know every screw in the model." White brick, black metal roofs, black-and-white marble, no-frills checking and savings, lollipops and dog biscuits — nearly every store was identical. Marketing chief John Cunningham: whichever store a customer walks into, they know exactly what to expect. Hill decided on locations himself — he also owned 45 Burger Kings in the Philadelphia suburbs — busy but not too congested, with the right mix of residential and commercial, and often right in a competitor's backyard. If a rival closed a branch, nearby Commerce employees got a $5,000 bonus. Payback was 12 to 18 months, versus three years for the industry. When it rained, staff walked customers to their cars with umbrellas; 300,000 pens were given away each month. Phones next to the ATMs dialed straight into the call center; online, you could see images of the front and back of your checks the next day. When Hill noticed other banks starting to refuse large amounts of coins or charge for them, he spent $10 million on the Penny Arcade: in 2001 it handled 750,000 transactions and took in $71.7 million in coins, free for customers and non-customers alike. Falese called it Marketing 101: once people come in, keep working with them. At a new store in Manhattan, meeting planner Carole Robbins had finished her banking by 8:15 a.m.; across the street was the Citibank that had swallowed her old bank. She said this was old-fashioned service, "like we're all in the same boat. But time will tell. Every restaurant is good at the start." Mergers were Commerce's deposit engine — whenever others merged, Commerce's deposits tended to jump.

WOW! was an operating system, not a slogan. In 1994, nine people wrote the SMART principles (Exhibit 9): say Yes to the customer, make everyone feel taken care of, keep your promises, know how to recover, think like the customer. Internal and external service quality were measured, and hitting the targets was celebrated. WOW! stickers could be exchanged for sweatshirts, mugs and radios. The WOW! Awards began in 1997; the most dazzling, "Retailer of the Year," came with a Porsche Boxster for a year; for the 30th anniversary in 2003 they planned to rent out Radio City Music Hall. Commerce University had 41 full-time staff. New employees attended Traditions: half games, half training, half common sense, teaching how to smile ("smile with your teeth showing"), shake hands and greet people. Hill told employees: in a sense you have joined a service cult. Hiring looked for people who would get addicted to giving great service; as John Manning put it, "we hire wack jobs." New hires were given a stack of cards to hand out whenever they saw someone giving good service at another bank, a restaurant or a gas station. There were Red Fridays, the WOW! Patrol, mascots Mr. C and Buzz, and a Dr. Wow whom nobody had ever seen, who replied to letters and gave out prizes. With Drexel, they ran Camp Business for high school students. Mystery shops (WOW! shops) were the hard core of evaluation: 5 employees managed 100 mystery shoppers doing 14,000 visits a year; staff had to shake hands, use the standard greeting (Hi, my name is ___. How may I help you today? — "can" could not substitute for "may"), walk customers to their seats, follow the manual, and act like they meant it. Raises depended on shop results; scores were posted on the intranet, and regions competed in leagues. They also opened accounts at competitors to test their service. Suggestions to "kill a stupid rule" earned $50 if adopted. At the call center you could easily escape the phone tree to reach a person, and agents' dress code was far stricter than the industry's — DiSabatina said you never know who might walk in: Hill, shareholders, the press. The vans were painted blue and red as free advertising; in New York some drove around empty. Hill said rivals didn't dare paint their vans for fear of being robbed, but the vans carried checks, not cash, and there hadn't been a robbery since 1945.

Hiring was always the bottleneck. For 40 positions in Manhattan they interviewed 2,000 people. Falese reviewed new hires at three months, six months and a year; in ten years, only one executive was found to be a bad fit within four months. He was willing to hire retirees part-time; what he wanted was thirty years of local experience — knowing which entrepreneurs could survive a cycle. Mergers created talent pools at competitors, and he had their org charts in hand, from loan officers to secretaries.

Retailtainment was the next step in 2002. Competitors had started copying the basic service: evening and weekend hours, ads playing up the personal touch. Washington Mutual (WaMu) came into New York as the "un-bank": roving tellers, kids' areas, no desks. The Dutch bank ING opened café-style branches in Manhattan, with espresso alongside savings accounts and free internet. Bank of America was experimenting with TVs for people waiting in line. Pauls said: we need the next penny arcade. Commerce deliberately didn't compete on price, using lower deposit rates to pay for better service. The question was how far service differentiation should go. A friend of Jacovelli's said: it was nice to be greeted and handed a pen, but she'd rather that person got behind the counter and moved the line along. Another customer complained of "too many greeters standing around." Some stores had already botched it with the hot dog cart and the juggler and ended up in the news. On one side: managers know their customers best and should be encouraged to delight them; on the other: total freedom is dangerous. Some thought it could work with training and guidelines; others thought attention and resources shouldn't be spent on it at all, and the existing service model should simply be executed well. Jacovelli's conclusion: Commerce is one bank, one brand, and its success depends on whether its service is better than its rivals'. The question is what role Retailtainment plays in that mission.

Key Data

Dimension	Figures / Facts	Source
Scale	2001 deposits $10.23B, assets $12.27B, 185 stores, 5,329 employees, net income $105.0M	Exhibit 1–2
Growth	Deposit CAGR 1998–2001 31.5%, 2001 +37.5%; households +20%; industry deposits 2001 +5.1%. Net income doubled vs industry +20%. Organic growth: 18 years to $1B, now can exceed $1B in a single quarter	Text p.4–5; Exhibit 2, 4; own calculation
Profit and cost	ROE 15.0% vs industry 12.4%. Efficiency ratio 70.3% vs 59.7% (more expensive). Cost of deposits 2.37% vs 5.35%. Securities 42.5% of assets vs 18% (fewer loans, more securities)	Exhibit 1–4; own calculation
Risk	Loan-to-deposit ratio 45.5% vs 87.1%; non-accrual loans 0.38% vs 1.44%. "Some of our best loans are the ones we didn't make"	Exhibit 2b, 4b; text p.6
Market and rivals	New York deposit pool $500B; WaMu un-bank, ING café, BofA queue TVs begin copying the surface; Commerce deposits jump when others merge (Exhibit 8)	Text p.4, p.11; Exhibit 8
Operations	7:30–20:00 + weekends + midnight drive-through + 10-minute rule; 4 checking products; self-built for $1M; payback 12–18 months vs 3 years; 40,000–100,000 visits/month; online 34%; Penny Arcade $10M / 750,000 transactions / $71.7M; 300,000 pens a month; 14,000 mystery shops/year	Text p.4–10
Site selection and promotion	Hill decides personally, often in rivals' backyards; $5,000 bonus when a rival closes; New York $500,000 promotion per store, 10,000 hot dogs, 4× the cost, half the payback	Text p.4, p.7
Incentives	Deposits and loans both credited to the store; no deposit, no loan; loan officer = store manager; raises tied to WOW shops; Kill the stupid rule $50; Retailer of the Year wins a Boxster for a year	Text p.6, p.8–9; Exhibit 10
Industry comparison	Cross-selling 1.5–2.5; attrition up to ~33%, best 15%; reasons for leaving 34% fees/service/errors / 34% moving / 15% convenience; satisfaction credit unions 79% vs banks 53%; number of banks 8,773→8,080, branches 61,957→65,654	Text p.2–3; Exhibit 3b, 5
Timeline	1973 founded → stock 20× after 1990 → 1994 WOW / Pauls joins → 1997 Awards begin → 2001 deposits $10.2B, big year for Penny Arcade → 2002 Retailtainment pilot and hot dog cart accident → decision point: should Jacovelli turn the Friday juggling into a system	Whole case

Key People

Person	Role	Position / role in the story	Relationships
Vernon W. Hill II	Founder, Chairman and CEO	Holds decision rights. Retailer logic: store, not branch; fight for the 62%, not the 3%; organic growth; a low expense ratio is draining capital. Co-owner of 45 Burger Kings in the Philadelphia suburbs; picks sites by instinct	The source for Jacovelli / Pauls / Falese; the antithesis of the industry
Deborah Jacovelli	Dean of Commerce University	The decision-maker in this case. She built the training and culture system; worries Retailtainment will undermine service consistency, and doubts whether customers even want to be entertained	Accountable to Hill's service mission; must define the pilot's role
Douglas Pauls	CFO	A conservative accountant proven wrong by the growth. Wants "the next penny arcade." Reminds everyone the ground war is in the stores — don't forget it	Gatekeeper of the numbers; feels the same pressure for a next step as Jacovelli
Dennis Falese	Retail/lending executive (spokesperson for the lending philosophy)	No deposit, no loan; stores get full credit; tight credit control; has every rival's org chart	Turned lending into a deposit-gathering tool
John Manning	Famous trainer	"Don't satisfy customers — blow them away"; hire wack jobs; the live persona of Traditions	The voice of the culture
John Cunningham	CMO	Stores look the same, so customers' cognitive cost is zero	Champion of brand consistency
Thomas DiSabatina	VP of Training and Development	The call center must WOW too; dress is a responsibility	Extended the culture down the phone line
Dennis DiFlorio	EVP of Retail Banking	"Don't come looking for a job — come to be a leader"	The hiring message
Carole Robbins	New customer in Manhattan	Fled her old bank after Citibank acquired it; likes Commerce, but leaves the line "every restaurant is good at the start"	External pressure: the honeymoon will expire
Rivals	WaMu / ING / BofA / Citibank, etc.	Starting to copy hours and the "personal touch"; mergers create lost customers and talent	Source of external pressure; also a source of Commerce's deposits and hires
Front-line managers	Each store	The executors of Retailtainment; know customers best, and are also the most likely to damage the brand	The layer Jacovelli doesn't dare fully empower

Hill holds strategic decision rights. Jacovelli has design authority over this pilot, but cannot single-handedly change Hill's retail religion. The money comes from low-cost core deposits, not loan spreads. Opposition to Retailtainment comes from inside the organization (just execute the existing model); outside, customers complain about idle greeters and lines that don't move.

Core Tensions

Retailtainment: use Friday improvisation to find "the next penny arcade," or execute the service model that already works all the way? (The professor's Q2 lands here.) On one side, rivals are starting to copy the basic service and Pauls wants the next wow; on the other, a hot dog cart on fire, greeters standing around, and Jacovelli worried about consistency. The case gives no success figures for the pilot — just one news-making accident and divided internal opinion.
Once you've grown up, can you still not be a "big bank"? Pauls's line on the cover. 185 stores, 5,329 people, 4× the cost in New York — standardization (the same screw) and local improvisation (managers doing whatever they like) are competing for the same brand.
Expense ratio 70% vs 60%: investment or waste? Hill says a low expense ratio is draining capital; the efficiency ratio is indeed 10 points higher, yet ROE still beats the industry (15% vs 12.4%) because the cost of deposits is about 300 bp lower. The trade-off holds as long as customers keep paying this price for convenience (accepting less interest).
Keeping people in the stores vs the industry pushing them online. Stores are expensive. Commerce deliberately made its stores easy to use, and online usage ended up higher (34%). This is channel complementarity, not channel substitution.
How far should service differentiation go? Umbrellas, lollipops and the Penny Arcade are designed, replicable wow; guitarists and burning hot dogs are unreplicable wow. That friend's call: she wants the line to move, not to be entertained.
Rivals copy the surface — should Commerce double down on the surface? WaMu's kids' areas, ING's espresso, BofA's TVs are all theater. What's genuinely hard to copy at Commerce: 4 products, the 10-minute rule, deposits credited to stores, 14,000 mystery shops, 40 chosen out of 2,000, no acquisitions.