The software industry has only two ways left: AI growth or high profits

Author: David George, Partner at a16z

Translation: Yuliya, PANews

Original Title: There Are Only Two Paths Left in the Software Industry: Grow 10% with AI Native Products or Achieve 40% True Profit Margin

PANews Editor’s Note: a16z partner David George released a new article stating that there is no “middle ground” in the software industry. Companies must choose between two paths within 12-18 months: achieve over 10% revenue growth through truly AI-native products, or boost real profit margins to 40-50%. He calls for a thorough restructuring, clear development direction, and reorganization of teams and architecture to adapt to the AI-driven new competitive environment, or face valuation compression and market pressure.

Below is the full translation:

To CEOs, founders, boards, and investors of software companies: the comfortable middle ground is over.

The public markets have re-priced the industry, and for good reason. The market is telling us that the terminal value of software has changed dramatically. I don’t know what factors will drive stock prices up or down next quarter, but in the medium to long term, I believe there are only two reliable paths to create lasting equity value.

• Path One: Accelerate revenue growth by more than 10% year-over-year within the next 12-18 months through genuinely new AI-native products.
• Path Two: Restructure the company to achieve a true operating profit margin of over 40% (ideally 50%), including stock-based compensation (SBC).

Strictly speaking, these measures are not mutually exclusive. But I believe this 12-18 month plan must be a binary choice. By the end of next year, any state between high growth and high profitability will become a no-man’s land: facing growth pressures, ongoing equity dilution, and valuation multiple compression. Today’s CEOs need clear actions, with one of these paths as the ultimate goal.

The game of adjustments is over

Publicly traded software companies have gone through the first half of their transformation. Growth has slowed, valuations have compressed. But in most cases, true profitability has yet to arrive.

Yes, free cash flow has improved; GAAP profit margins have increased. But once you treat stock-based compensation as a real expense rather than a permanent exemption, most companies remain stuck in a tough middle ground: growth is too slow to deserve high-growth valuation premiums; equity is too diluted to command stable valuation multiples.

If revenue growth is slowing, we should see more operating leverage, but while we see some, it’s still far from enough.

The reality is, now is the time for management to take bold action. And those headlines about “laying off 8% or 10%” no longer work. That’s just weak performance. Weakness is just trimming organizational edges without addressing core issues. Stronger measures involve redesigning and restructuring the entire organization and operational model.

In the next 12 months, I expect to see more tough measures. And how you do this depends on how you want to reorganize your company.

Path One: Accelerate growth with new AI products

Accelerating growth with new AI products doesn’t mean adding chatbots or Copilot interfaces to your existing SKU list.

It means launching new products within 12 months that can boost your company’s total growth rate by 10 percentage points. Equally important, it requires restructuring your company—especially your executive team—at the fastest speed possible to ensure that once product-market fit (PMF) is found, you can quickly seize market opportunities and hit your growth targets.

The first step is identifying who will be the leaders helping you accomplish this. It will be a tough 12-month journey, and you need to find those willing to go through it with you. But there’s good news: within your organization, about five people will bring you 100 times the value you expect. Your primary task is to identify who they are (regardless of their background), explain the urgency of the situation, and offer them a once-in-a-lifetime career opportunity to help you restructure the company.

How should you arrange these people?

Start by assigning them to critical but overlooked information-gathering projects:

• Conduct process capture sprints around high-value workflows;
• Collect SOPs, work orders, chat logs, requirement documents, policies, CRM notes, support logs, event data, and approval paths.

Create a dynamic context layer instead of a bunch of static PDFs. Treat documents as foundational infrastructure for your products. Establish evaluation mechanisms around accuracy, exception handling, latency, and cost. Immediately involve these five individuals, each with their own scope of responsibility.

Over the next month, closely monitor your vice presidents to see who is aligned with that team and who isn’t.

This will tell you everything you need to know: which executives should stay and which should part ways in the upcoming restructuring.

At the end of the month, have conversations with those vice presidents and directors who need to leave. Replace them with the elite team that just completed the information collection sprint, along with other proven AI-native rising stars within the company.

Now, you have a revitalized, energized executive team ready for battle.

Meanwhile, allocate 50% of R&D resources to new AI products.

Use a four-person squad model; merge design, product, and engineering into one working unit, start coding from day one, limit personnel rather than compute power. Minimize communication overhead.

Ensure your top product managers spend as much time as possible directly engaging with customers. They shouldn’t waste a minute. Their job is pure product exploration; they must be unencumbered by legacy issues.

At the same time, your best engineers should stay within the central engineering organization, reporting directly to the CTO. Their role is to ensure the core engineering architecture evolves as quickly as the pioneering product managers.

Different companies may have different circumstances, but my advice is: don’t place all your top engineers on the periphery. It’s tempting, but it will fragment your tech stack and create years of technical and organizational debt, killing any early promising progress.

Additionally, in AI, you don’t need the absolute best engineers for exploring new products; you need those who can deliver and learn quickly. The top engineers should keep an eye on the overall company architecture but prioritize new initiatives ruthlessly.

As part of this sprint, your company must be very adept at upgrading controversial decisions to clear obstacles. If you can’t make tough choices weekly, you won’t complete this transformation in 12 months or successfully build a new AI-native business. So master this process and ensure your newly formed leadership team dedicates a significant portion of their time (at least one full day each week) to removing obstacles for designers, product managers, and engineers—as if the company’s survival depends on it.

In the process of clearing obstacles, you will precisely understand what your new business model is. It needs to generate revenue via tokens/per-use rather than the old user-based payment model. You still have some time: user-based pricing won’t disappear overnight. But you must take this challenge seriously: don’t be sloppy with new pricing models and product interfaces. If agents can’t autonomously use and pay for your products, you probably haven’t reached your goal.

The budget for new expenditures exists. You can do this.

But remember, the most immediate and obvious source of AI savings for your customers is labor efficiency, meaning seats will be where they seek to cut costs. In contrast, new growth will increasingly come from tokens, consumption, automation, results, and machine-driven workflows.

If you’re not on the token path, you’re not in the fastest-growing part of the budget.

Not all companies are positioned to do this. You might evaluate your options and see no reliable hope of winning via Path One. But if you do see it, and if you endure this twelve-month sprint, you will stand out as a focused and accelerating company with a new leadership team and a “reboot moment,” from which your team will gain unity and new vitality for years to come.

Path Two: Restructure for over 40% true profit margin

Over the past decade, software companies have been very good at talking about free cash flow margins. But if we take this seriously, we should stop excluding equity incentives and pretending that dilution isn’t a cost borne by shareholders. For those not planning to re-accelerate growth, I believe the right goal is to achieve a true operating profit margin of 40% or even 50% within 12-24 months (including SBC).

Reaching over 40% profitability requires more than 10% or 20% layoffs. It means flattening management layers, standardizing implementation, minimizing custom services, eliminating committees, raising prices where workflows or switching costs have an advantage, shifting long-tail customers to higher baseline prices or letting them churn, and viewing every issued share as a transfer from owners to employees.

AI should reshape the company’s form. Cost structures should change accordingly.

This effort will be similar to Path One. Even if your goals differ, you still need to build an AI-native company within 12 months, maximizing engineer productivity and efficiency. From day one, you must envision what a smaller, more motivated, and more productive team looks like after twelve months.

Counterintuitively, the first thing you should do is significantly increase the token budget allocated per engineer. If your engineers aren’t spending real money on tokens, they might not be pushing hard enough. A monthly $1,000 per engineer isn’t excessive; it’s almost a baseline requirement.

A useful premise is that the upper limit of output per engineer can grow much faster than most organizational structures can leverage. Some top operators have described seeing an order-of-magnitude productivity increase among top engineers managing 20 to 30 agents. Whether 20x is extreme or just cutting-edge, the impact on the organization is the same: a company built for ten people in speed will outperform one built for four.

Meanwhile, prepare for large-scale layoffs—you already know this.

You can’t just trim the edges: if you cut a large portion of your independent contributors but keep the director and VP teams, your situation will worsen. It’s important to note, this differs from Path One; you’re not trying to build a “new” business. But you are “re-creating” the company around a new set of values—performance and shareholder mindset—so ensure you’re on this journey with the right leadership.

Another critical point: your team must honestly assess which old moats are being eroded.

Having data alone is often not enough.

Integration becomes easier to copy.

As agents can more easily move across systems, workflow and UI advantages matter less. Migration becomes simpler.

Competitors will increasingly attack each other’s core modules, not just edges. This means core business pricing pressure is imminent, so prioritize advantages that help you maintain pricing power and customer retention.

This is achievable: lessons from Broadcom

Before AI, there was a case study of a tough approach in the public markets: Hock Tan’s leadership at Avago/Broadcom (*Note: In 2013, Avago acquired LSI for $6.6 billion, entering enterprise storage; in 2016, Avago acquired Broadcom and renamed itself Broadcom Limited). It’s a brutal model. It’s not every founder’s cultural blueprint. But it reminds us that aggressive cost control, product simplification, and price realization are possible. Tough measures do exist.

The second path may sound somewhat pessimistic, but not every software company has the right to choose Path One. If a company doesn’t have that right, then Path Two is the only way to create value.

Key questions

Founders should write a question on the first slide of every board presentation: Which path are we on?

Is it through new AI products achieving over 10% revenue growth? Or reaching over 40%+ true operating profit margin including SBC?

Investors should ask the same question more forcefully now.

Where is the AI engine capable of changing the curve? Where is the restructuring of R&D around small, token-rich, customer-close teams? Where is the plan to build human/agent dual interaction layers? Where is the clear roadmap to achieve 40-50%+ true profit margins? Where is the plan to reduce equity dilution as a proportion of revenue?

If the answer is some version of “both with a bit of each” or “we’re evaluating options,” I expect the market to continue applying pressure.

Founders: you need to pick a path, and you need to quickly decide who in your team you want to go with you. You have the opportunity to create a new entrepreneurial moment for your company, your new team, and your investors. Grow 10%, or earn 40%. Build the next-generation product, or build a cash cow. No middle ground—good luck.