Imagine a zoo that has bred its own animals for generations. Every inhabitant has lived in the protected zoo environment its entire life, as did its parents, and its parents’ parents. There is no knowledge of life outside the zoo.

All at once, the animals are released into the wild.

Chaos ensues. The zebras don’t know which grasses to eat. Yaks are gnawing on flamingos. Predators stroll right up to naive herbivores and devour them.

This is basically what has happened to the American (and Australian, and some of the European) populace over the past 50 years with retirement savings. Once the province of corporate pensions and defined benefit retirement plans, the onus of adequately saving and investing for retirement has been shifted to the individual’s own shoulders. In most cases, those shoulders are wholly unprepared to bear the burden. (Except Investing Daily members, of course, who have met this challenge with gusto.)

Most investors in this new Darwinian financial jungle are prey. People aren’t born with investing survival skills, and most lack the time, interest or aptitude to gain them.

Amidst the chaos, in strode Financial Engines (NSDQ: FNGN). Backed by Nobel Laureate William Sharpe (best known for the Sharpe ratio, an elegant but oversimplified way to judge a fund’s performance), Financial Engines stepped in to provide desperately needed 401(k) advice and management. To American families, the company promised to supply the investing expertise once filled by pensions. To employers, Financial Engines promised more financially healthy employees at a fraction of the cost. And it promised a business model that would scale beautifully, using technology rather than human advisers to dispense tailored advice to millions.

Much like the Sharpe ratio, though, the idea is better on paper than in practice.Twenty years in, Financial Engines is not the business investors hoped it would be. That’s unlikely to change. Yet the stock is still priced like a high-growth tech start-up, a valuation that comes with expectations that Financial Engines cannot meet. As it inevitably falls short, the stock should experience a well-deserved shellacking.

The Promise

Financial Engines’ investor pitch is compelling. The company has partnered with the eight major 401(k) providers dominating that industry, giving it access to 693 major U.S. employers and their 9.3 million employees. Once advising an employee on a 401(k) plan, Financial Engines can expand its advice to non-retirement investments. That relationship can continue after the employee retires, too.

To grow, Financial Engines could partner with new 401(k) providers and advise more  employees. Because fees are a percentage of assets, revenue grows simply as employees save more and as the market rises over time.

All the advice is developed using computer programs. Financial Engines studies a company’s 401(k) plan, writes a bit of software to customize advice for any given employee, and voila! The same process applies to non-retirement plan investments, which should be even easier to set up without the constraints of a fixed 401(k) menu. In theory, this creates scale: Once written, the programs should cost basically nothing, whether they serve the second, third or thousandth employee of each company.

The Reality

The results tell a different story. Rather than beautiful scale, costs have risen almost perfectly in tandem with revenue.

Far from shrinking over time, costs are roughly flat, even increasing last year. Why?

As it turns out, slapping an algorithm (program) on a 401(k) plan is a fraction of the company’s costs. Financial Engines still needs to sell plan providers (the companies that administer the plans), employers and employees on paying for the service. It’s been successful with providers, less so with employers, and much less so with employees. Twenty years in, only 10% of eligible employees are customers. That percentage is climbing, but painstakingly. Financial Engines currently spends $1.60 in marketing for each incremental dollar of revenue. Considering 84% of that dollar is eaten up in expenses (the exact same proportion as five years ago), the economics are less appetizing. Convincing customers to extend their advice to outside investments or stick with Financial Engines after they retire has been harder still.

Also, with half of eligible customers having less than $55,000 in their 401(k), managing millions of small accounts isn’t as efficient as running a handful of multimillion-dollar accounts. For example,  customers—surprise!—often want to speak with an adviser.

Not that the company hasn’t grown. Revenue has climbed an average annualized 24% over the past five years, though growth has slowed each year (down to 9% last year) and costs have risen at the same rate.  At $115 billion under management, Financial Engines is bigger than the next seven competitors combined.

The low-hanging fruit has been plucked, though. With every major 401(k) provider already a partner, growth must now come from convincing more employees to sign up. Not only is this Financial Engines’ weak spot, but this growth is harder to come by. Truly tech-first robo-advisers, including Wealthfront, Betterment and Future Advisor, are swooping into the 401(k) market with better technology, higher conversion rates and profitability that grows with the business.

To its credit, Financial Engines’ management seems to see the writing on the wall. By acquiring The Mutual Fund Store—yes, a brick-and-mortar fund company—and hiring call centers and human advisers for higher-touch service, Financial Engines is recognizing that its place in the 401(k) market is stodgy incumbent, not nimble disruptor.

But the market shouldn’t pay 65 times earnings for a stodgy incumbent. That price made sense only when Financial Engines was on its moon leap. My model generously values the stock at $19 per share, which means it has more than 40% to fall as gravity takes hold.

Logistics and Options

Financial Engines’ stock is plenty liquid, and shares should be easy to borrow. Interactive Brokers, my personal broker, puts the cost to borrow at just 1.05%. Check what your broker’s rate is for the stock. If it’s above 7%, consider the option alternative below or, frankly, switch to a better broker.

Options on the stock only go out to March, so I recommend shorting the stock directly. If you’d like, though, you can set up a synthetic short. With shares in the low $30s, simultaneously go long the March $30 puts and sell (short) an equivalent number of March $30 calls. The net amount you’ll pay to set this up (the premium you pay for the puts less the premium you receive from calls) should equal a bit less than the share price at that moment minus $30. The result of this trade will be a payoff profile identical to that of shorting the stock, except that your trade expires in March, so you’ll need to “roll” it, or set up a new version, as expiration approaches.

Financial Engines pays a 0.88% dividend. While we’re short, we’ll be paying that dividend to the folks from whom we borrowed shares. This is a small price to pay for 40% upside. A note, though, because this is a common mistake I see: You cannot avoid paying the dividend via a synthetic short. The dividend yield is imputed in the options prices; that’s why the net cost to set up the trade is a bit less than the share price minus $30 rather than exactly equal to it.

Alex Pape, CFA, manages accounts that short Financial Engines’ stock.