Great strategy? Too bad, you fail.

In 1996, General Motors did something no major automaker had done before. It built a real electric car. Not a concept. Not a prototype sitting behind glass at an auto show. The EV1 was road legal, aerodynamic, and genuinely loved by the people who drove it. On the best battery it offered, it went about 140 miles on a charge, respectable for 1999, unheard of for anything GM had built before.

GM never sold a single one. Every EV1 was leased, a little over eleven hundred of them, to fewer than a thousand customers in California, Phoenix, Tuson, and Atlanta. There was a waiting list of thousands more who wanted in.

Seven years later, GM crushed almost every one of the vehicles.

Here's the number that should bother you. The whole EV1 program, from early development through shutdown, cost GM roughly a billion dollars over about a decade. Call it a $100 million a year. Against $186.8 billion in annual revenue, that's a rounding error, a nickel out of every hundred dollars GM took in. This wasn't a company forced to choose between the EV1 and the truck lineup because it could only afford one. There was no real tradeoff. GM had the money to keep both. It chose not to.

Years afterward, GM's own CEO called killing the EV1 his worst decision. He wasn't talking about bad timing or a market that wasn't ready. He was owning a strategic error. A legacy company the size of GM needs a place to protect promising bets like the EV1, cheap insurance against a future it can't yet prove, insulated from the quarterly scorecard the truck business gets judged on. GM never built that protection. It let a $100 million bet get judged by the same standard as a business generating tens of billions of dollars. By that standard, of course the EV1 lost.

A decade later, the company was racing to catch up to an industry it had once led by ten years.

This is what most conversations about strategy get backwards. We assume failure means the decision was wrong. Here, the decision was right and the company still killed it, not because it had to, but because no one ever decided the EV1 deserved to be protected from a fight it was never built to win. GM gained little nourishment by eating its seedcorn.

A bad decision isn’t the starting problem

The problem is a good decision that isn't happening. That’s the Execution Gap. It’s why good strategy doesn’t get done.

The distance that matters isn't always between one strategy and a better one. It's often between what we decide we will do and what actually gets done. Deciding on a strategy isn't the same as executing it. "We're going to grow sales" or "We're launching the new service" are choices. They are not plans.

GM didn't fail to decide. It failed to protect a decision that never should have been up for debate against the truck lineup in the first place.

Five Places Execution Breaks

After years of watching this play out in board rooms and strategic planning sessions, I've come to see execution failing in the same five places, over and over. Ownership. Translation. Priority. Resources. Feedback. Miss any one of them and the strategy stalls, no matter how sound it was on the day you decided it.

Gap 1: Ownership

Listen for these phrases. "We need to." "Someone should." "The team is working on it." Each of them is a warning sign. If everyone owns something, no one does.

There's a sharp distinction between activity and accountability. "Alex is doing the marketing" describes activity. "Alex owns generating ten qualified leads per month" describes a result someone is accountable for. The diagnostic question is simple: Who is accountable for the result rather than the effort?

GE learned this the expensive way. Its thesis was that every jet engine, turbine, and MRI machine it built could be embedded with sensors, feeding data into software that predicted failures, optimized performance, and created new revenue streams. Pull it off, and GE wouldn't just be selling machines anymore, it would be selling the software layer running on top of them, competing with the likes of Microsoft and Oracle. It was a coherent, well-capitalized bet on where industrial equipment was headed.

GE built the strategy inside a separate division called GE Digital, essentially its own company within the company, with its own leadership, its own budget, headquartered in San Ramon, California, far from GE's industrial businesses. GE Digital would build the software. The product divisions - aviation, power, healthcare - would then embed it into how they built and sold their machines. THe problem was that nobody made that adoption mandatory, and nobody made the industrial divisions accountable. The leaders running GE's jet engine business and its power turbine business still had their own targets, budgets, and incentives, all tied to selling and servicing physical equipment the way they always had. GE Digital had no authority to make the divisions change that. And those division leaders faced no consequence if they didn't.

The bill came due. GE spent more than $4 billion building the digital business, including a $915 million acquisition of ServiceMax in 2016 to extend the platform into field service. Two years later, GE sold 90% of ServiceMax at a loss and spun the rest of the unit off into a standalone subsidiary, quietly retreating from the ambition it had spent billions chasing. GE Digital's CEO departed at the end of 2018. By 2020, the company had stopped disclosing separate results for the business. The top-ten software company target had a deadline of 2020. What it got instead was a write-off.

GE had built a capable software business, and at the same time had no one whose job depended on that software actually getting embedded across the products that mattered. GE Digital owned building the technology. Nobody owned making the rest of the company use it.

Gap 2: Translation

Strategy states intent. Execution requires defining what happens next. "Grow recurring revenue" is a strategic choice. It rules things out, prioritizing repeat revenue over one-time sales, but it doesn't tell anyone what to do on Monday morning. Someone could translate this strategy into moving existing customers onto an annual contract. Someone else could translate it just as faithfully into a membership tier, or an auto renewal option on the core product. All three serve the strategy. None of them swaps in a different goal. That's what makes this a translation problem and not a strategy problem: Multiple faithful, competent translations of the strategy exist and would work. What is best in the bigger picture is not clear.

This is where it gets harder than picking an option off a menu. An individual or a group can translate a strategy on their own into a reasonable action or set of actions…and still get it wrong, That’s because they're translating with a narrower view than the organization may need. They may not know that finance already flagged limited capacity to service new subscription billing this year. They may not know that the product people are in mid build of a feature that would conflict with a membership tier. They may not know that leadership is deliberately avoiding new contract types because a related initiative already strained the operations team. Their translation may be faithful to the stated goal and still be wrong, because faithful to the goal and coherent with everything else the organization is doing are two different tests. Only someone with visibility across the strategic plan and the organization can check both.

A translated strategy answers what exactly happens, who does it, by when, in what sequence, and what done looks like. It answers those questions in a way that fits in with everything else that has already been committed. The test: If people have to interpret the strategy before they can act on it, or if what they pick could clash with a strategic decision made somewhere else in the business, the strategy has not been properly translated yet.

Boeing found this out with the 787 Dreamliner, its mid-size, long-range jet built largely from carbon-composite materials instead of traditional aluminum to cut weight and fuel use. It became the fastest-selling commercial jet in aviation history on orders alone. To build it, Boeing broke from its tradition of manufacturing planes mostly in-house near Seattle and instead outsourced roughly 70% of the aircraft's design and manufacturing to a global network of about 50 partners.

The strategic call wasn't reckless. Airbus does something similar in outsourcing manufacturing. The problem was that Boeing handed each of those 50 partners the concept and let each one translate it on its own, with visibility into its own piece of the plane and little else. Suppliers were often given incomplete specifications and had to fill in the design work themselves. The suppliers’ individual choices weren't unreasonable. But nobody was translating the strategy with a view of how all 50 suppliers needed to sequence and integrate on one timeline. Locally faithful, globally incoherent: That mismatch is what strung the program out, with the result that the plane appeared nearly three years late and it was billions of dollars over budget.

The outsourcing strategy wasn't the failure. Not coherently translating it for the 50 suppliers was the failure. The strategy had to be translated 50 separate times, with different results. There was no coordinated, big picture provided.

Gap 3: Priority

Calendars and budgets tell you the real priorities. The strategic plan on the wall says what you meant to prioritize. Your calendar and your budget say what you actually are doing. "10 priorities" means no priorities. Calling something “important” forces a harder question: What will you stop, postpone, delegate, or say no to in order to make room for it?

People often miss this: Sometimes there's nothing to stop, postpone, or delegate, and the priority still doesn't survive.

GM's EV1 cost roughly a $100 million a year against a $186 billion in revenue. There was no real competition for scarce dollars. GM hallucinated a tradeoff that didn't need to exist, then let the imgained tradeoff kill a program with a waiting list of buyers. That's a priority failure of a specific kind: No one ever decided the EV1 deserved protection from being judged by the same standard as the core business. A skunkworks, innovation bet judged on this quarter's numbers will lose to trucks every single time, whether it costs $100 million or $1 billion, because it isn't playing the same game.

Here's the part that makes this harder to explain away. California's zero emissions mandate is why the EV1 program started, but GM didn't stop at California. Within the first year, leasing expanded to Arizona and Georgia, when neither one had a regulatory requirement to sell clean vehicles. GM went there voluntarily, working with local utilities to study real-world charging and demand outside the one state actually forcing its hand. That's not the behavior of a company doing the bare legal minimum.

And the people closest to the car weren't lukewarm either. Chelsea Sexton, who sold and serviced EV1s from inside GM starting in 1996, described sales reps being told the credibility of 400 people and 10 years of work rode on them, and said flatly that front-line staff gave management meeting after meeting of customer feedback, only to watch "exactly the opposite of our recommendations happen." Her account points to weak, inconsistent advertising as a recurring complaint from a sales team that believed in the product. Her own words for what was actually happening inside GM: "There are factions in GM." Not one company making one calculated call. Multiple groups pulling in different directions, with the people who wanted the EV1 to win losing to the people who didn't.

That reframes the failure. This wasn't a case of nobody caring enough to fight for the program. The EV1 had real engineers, sales reps, and customers with a waiting list wanting it to succeed, But it still died. The gap wasn't belief. Instead, no one senior enough insulated the EV1 from being judged by the truck lineup's standards, no one made its budget non-negotiable, no one had the authority to keep saying “yes” once the regulatory pressure that had forced GM's hand in the first place eased up in the late 1990s.

GM's CEO Rick Wagoner later called killing the EV1 his worst decision. Read against Sexton's account, that sounds less like regret over a close call and more like an admission that the company let something people genuinely wanted and that was important to GM’s future die anyway. Caring about a result and being accountable for protecting it turned out to be two very different things.

Gap 4: Resources

Every strategy assumes you have enough time, money, people, skill, technology, information, and relationships to pull it off. Calling something a priority doesn't resource it. Without deliberate resourcing, execution runs on spare time, extra effort, and luck, and usually none of those hold up for long.

In 2005, activist investor Carl Icahn built a stake in Blockbuster, won a proxy fight for three board seats, and immediately went after CEO John Antioco for what he saw as spending too much on the company's mail delivery service instead of the physical stores. Two years later, in 2007, Blockbuster launched Total Access anyway, combining mail rental with free in-store exchanges, an advantage Netflix couldn't touch. It worked. Blockbuster crossed two million online subscribers, fast enough that Netflix had to cut its own prices to compete.

Then Blockbuster's debt and Ichan’s pressure caught up with the strategy before it could fully play out. The company was carrying over $1 billion in debt from its earlier spinoff from Viacom, and its credit agreements required it to hit specific earnings targets to stay in good standing with lenders. Sustaining an aggressive, money-losing subscriber push against Netflix put those targets at risk. Between the credit covenants and Icahn's continued Board pressure, Blockbuster pulled back spending on Total Access in 2008, right as it was gaining ground. The idea had already proven itself. It just ran out of runway and resources before it could win.

Gap 5: Feedback

"We're working on it." "We've been busy." "We've made progress." Progress toward what?

Feedback is the gap that hides the other four gaps, because effort feels like proof of execution even when nothing is actually moving.

The fix is to measure movement instead of effort. Calls made becomes qualified conversations. Posts published becomes leads generated. Meetings held becomes decisions made and actions completed. If you can't answer what the number is supposed to prove, you're measuring effort, not results.

Peloton offers a recent example of this failure. Its strategy, premium connected fitness hardware bundled with a recurring content subscription, was never the problem. It's still a real business today. The strategy worked. What Peloton watched to judge whether it was working didn't.

During the pandemic, Peloton's headline number was connected fitness subscriptions, and that metric looked extraordinary. Growth of 87% one year, still climbing 27% even in the quarter everything started to unravel. Peloton kept expanding manufacturing capacity and building inventory to match that curve, treating subscriber growth as the number that mattered. But a subscription only counted that someone had bought a bike, not that they were using it, and that's exactly where the real signal was sitting unwatched. Average workouts per subscriber peaked at 26 a month in early 2021, then fell 43%, to under 15 a month, as people went back to gyms and offices. Churn nearly doubled over the same stretch. Peloton kept manufacturing and shipping against subscriber projections built on the pandemic curve, not against the engagement numbers quietly showing that a large share of new subscribers weren't sticking around.

By mid-2022, Peloton was sitting on far more inventory than it could sell,. It posted a $1.2 billion quarterly loss. Founder and CEO John Foley was forced out along with thousands of employees. “Subscriptions added” was an effort number. It counted commitments made. “Workouts per subscriber” was the movement number. It measured whether the product was actually delivering enough value that people would keep paying for it. Peloton had both numbers the entire time. It just kept building its strategy around the one that looked good.

The Execution Chain

Strategic choice leads to a desired result. The desired result needs an owner. The owner needs next actions. The actions need resources. Progress needs a measure. And the measure needs a review. Break any link in that chain and execution weakens, even if every other link is solid.

Here’s an example of what this looks like in practice.

Strategic choice: Develop the yacht club market.

Desired result: Five qualified yacht club opportunities by October 31st.

Owner: One named person.

Actions: Build the list, identify the decision makers, create the offer, contact them, follow up.

Measure: Qualified conversations, proposals, sales.

Review: Weekly.

Notice what isn't in that list. There's no ambiguity about who's responsible, what counts as progress, or when someone checks in. That's the difference between a strategy and an executable plan.

Your Turn

Pick one important strategic decision that you have made in your business that isn't moving. Ask which of the five gaps, ownership, translation, priority, resources, or feedback, is doing the most damage. Don't try to fix all five at once. Find the one change that would most improve execution and start there.

Here's a structure I use with clients that I call the thirty day commitment.

• Define the result.

• Name the owner.

• List the next actions.

• Commit the resources.

• Set the measure.

• Set the review cadence.

• Then execute.

This isn't another plan. It isn't another list. It's one result, one owner, and thirty days of focused execution.

Strategy determines where you want to go. Execution determines whether you get there. Your strategy is not what you say is important. It's what you consistently turn into action and results.

GM's own electric trucks now beat the range the EV1 could never get past, by three times over. The strategy caught up. It just took thirty years longer than it needed to.

What's the one result you'll refuse to let drift over the next thirty days? Write it down.

Reflection

I have been writing about business for more than 50 years, and I never tire of stories like these. There is something captivating in watching organizations with far more talent, capital, and resources than any of my clients will ever have blunder in exactly the ways smaller companies do, just with more zeros attached. GM, GE, Boeing, Blockbuster, Peloton: These are not companies short on smart people. They had strategists, consultants, boards, and budgets that most businesses could only dream of. And they still lost the thread between deciding and doing.

These much larger brethren keep making these mistakes in public, documented in annual reports, CEO interviews, documentaries, and news and feature stories, where anyone willing to look can see where the gap between strategy and execution opened up and what it cost. The lesson to be learned from GM crushing EV1s is sitting in plain sight, already paid for by GM’s billion dollar mistake.