The software industry is left with only two paths: either AI-native products grow by 10%, or real profits reach 40%.

Author: David George , Partner at a16z

Compiled by: Yuliya, PANews

PANews Editor's Note: David George, a partner at a16z, published a new article stating that there is no "middle ground" in the software industry. Companies need to choose between two paths within 12-18 months: achieve revenue growth of over 10% through AI-native products, or increase their true profit margin to 40%-50%. He calls on companies to undergo a complete restructuring, clarify their development direction, and reshape their teams and structures to adapt to the new AI-driven competitive environment; otherwise, they will face valuation compression and market pressure.

The following is the full translation:

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

The public market has already repriced the industry, and for good reason. The market is telling us that the end-user value of software is no longer what it used to be. I don't know what factors will drive the price movements of each stock next quarter, but in the medium to long term, I believe there are only two reliable paths to creating lasting equity value.

• Path 1: Achieve revenue acceleration of more than 10 percentage points year-on-year within the next 12 to 18 months through truly new AI-native products.

• Path 2: Restructuring the company to achieve a real operating profit margin of over 40% (ideally 50%), which includes stock-based compensation (SBC).

Strictly speaking, these measures are not mutually exclusive. However, I believe this 12- to 18-month plan forces a choice between two paths. By the end of next year, all states somewhere between high growth and high profits will become uncharted territory: facing growth pressure, continued equity dilution, and valuation compression. Today's CEOs need clear initiatives to pursue one of these paths as the ultimate goal.

The trick of adjustment is over.

Publicly listed software companies have already completed the first half of their transformation. Growth has slowed, and valuations have compressed . But in most cases, true profitability has yet to materialize.

Yes, free cash flow has improved; GAAP margins have also increased . However, once you consider stock-based compensation as a real expense rather than a permanent exemption, most companies in the industry remain in a difficult middle ground: too slow to warrant a valuation premium for high growth; too dilutive to warrant a stable valuation multiple.

If revenue growth is slowing, we should see more operating leverage, but again, while we have seen some, it is far from enough.

The reality is that now is the time for management to take bold action. And those headlines about "8% or 10% layoffs" are no longer effective. That's just a sign of weakness. A weak approach merely trims the edges of the organization without addressing the core issues. A stronger approach, in contrast, is to redesign and restructure the entire organization and its operating model.

Over the next 12 months, I expect to see more aggressive approaches. Regarding how to do this, you have two options, the difference being how you want to restructure your company.

Path 1: Accelerate growth through new AI products

Accelerating growth through new AI products does not mean attaching chatbots or Copilot interfaces to an old SKU list.

This means launching a new product within 12 months that can boost the company's overall growth rate by 10 percentage points. Equally important, it means you need to restructure your company as quickly as possible, including your executive team, to ensure that once you find product-market fit (PMF), you can quickly seize market opportunities and achieve growth targets.

The first thing you need to do is identify who will lead you through this task . This will be a grueling 12-month journey, and you'll need to find out who's willing to share the hardships with you. However, there's good news: within your organization, there are approximately five people who will bring you 100 times more value than you imagine. But your primary task is to identify these five individuals (regardless of their backgrounds), explain the urgency of the situation to them, and offer them a once-in-a-lifetime career opportunity to restructure the company with you.

How should you arrange these people?

First, put them in charge of those seemingly insignificant but crucial information gathering projects:

• Perform a process capture sprint around each high-value workflow;

• Collect SOPs, work orders, conversation logs, requirements documents, policies, CRM notes, support logs, event data, and approval paths.

Create a dynamic context layer, not a bunch of static PDFs. Treat documents as product infrastructure. Establish evaluation mechanisms around accuracy, exception handling, latency, and cost. Immediately get these five people involved in this task, each with their own responsibilities.

Over the next month, keep a close eye on your VPs to see who's on the same team and who's not.

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

At the end of the month, hold talks with the VPs and directors who need to leave . Replace them with elite teams that have just completed their information gathering sprint, as well as other proven AI native talents within the company.

Now you have a refreshed and dynamic executive team, ready to take on the challenge.

At the same time, you should invest 50% of your R&D resources in brand new AI products.

Employ a four-person team model; combine design, product, and engineering into a single work unit , start coding from day one, and limit the number of people rather than computing power. Reduce communication costs to as close to zero as possible.

Make sure all your best product managers are directly engaged with customers as much as possible. They can't waste a single minute. Their job is pure product exploration; make sure they're not held back by any legacy issues.

Meanwhile, your best engineers will remain in the central engineering organization, reporting directly to the CTO. Their job is to ensure that the company's core engineering architecture evolves as rapidly as the pioneering product managers.

The situation may vary from company to company, but my advice is not to marginalize all your best engineering talent. It may be tempting to do so, but it fragments your technology stack and creates years of technical and organizational debt, thus stifling any promising early progress.

Furthermore, in the field of AI, you don't need the best engineers to explore entirely new products; you just need people who can deliver and learn quickly . The best engineers should maintain a focus on the company's overall technical architecture, but ruthlessly prioritize new things.

As part of this sprint, your business needs to be exceptionally adept at escalating controversial decisions to remove obstacles to progress. You won't be able to complete this transformation and successfully build a new AI-native business within 12 months if you can't make tough choices every week. Therefore, master this process and ensure your newly formed executive team dedicates significant time (at least one full day per week) to clearing obstacles for designers, product managers, and engineers, as if the company's very survival depends on it.

In clearing obstacles for your team, you'll figure out exactly what your new business model is . It needs to monetize through tokens/pay-per-use , not the old pay-per-user model. You do have some time: pay-per-user models don't disappear overnight. But you need to take this challenge seriously: you can't skimp on the new pricing model and product interface. If agents can't use and pay for your product on their own, then you're probably not on your way.

The budget for new spending exists. You can do that.

But remember, your customers' first and most obvious source of AI savings is labor productivity, meaning that seats will be where they look to cut costs. In contrast, new growth will increasingly be seen in tokens, consumption, automation, results, and machine-driven workflows.

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

Not all companies are in a position to do this. You might evaluate your options and see no credible hope of winning through Path One. But if you do, and if you survive the 12-month sprint, you will stand out as a focused and accelerating company with a new leadership team and a “moment of reinvention” from which your team will gain unity and renewed energy for years to come.

Path Two: Restructuring to Achieve a True Profit Margin of Over 40%

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

Achieving profitability of over 40% requires more than just 10% or 20% layoffs. It means flattening management hierarchies, standardizing implementation, minimizing customized services, eliminating committees, raising prices where there are workflow or switching cost advantages, moving long-tail customers to higher floor prices or letting them churn, and treating every share issued as a transfer from owners to employees.

AI should change the form of companies. Cost structures should change accordingly.

This will require a similar level of effort as the first path. Even if your goals are different, you still need to aim to build an AI-native company within 12 months, maximizing the productivity and efficiency of your engineers. From day one, you need to figure out what a smaller, more motivated, and productive workforce will look like in your organization twelve months from now.

Counterintuitively, the first thing you should do is significantly increase the token spending budget allocated to each engineer. If your engineers aren't spending real money on tokens, they might not be pushing hard enough. A thousand dollars per engineer per month isn't excessive; it's practically a basic requirement.

A useful premise is that the output ceiling for an individual engineer increases far faster than most corporate organizational structures can utilize. Some of the best operators have described top engineers seeing orders-of-magnitude productivity gains while simultaneously managing 20 to 30 agents. Whether 20 times is an extreme case or merely the cutting edge, the impact on the organization is the same: a company built for a ten-person committee will lose in speed to a company built for a four-person commando team.

At the same time, you need to prepare for large-scale layoffs—you already know that.

You can't just prune the leaves at the company's edge: if you cut a large portion of the company's independent contributors but keep your director and VP team, you'll be in a worse position than you started. To be clear, this is different from Path One; you're not trying to build a "new" business. But you're still "re-establishing" the company around a new set of values ​​centered on performance and shareholder mentality, so make sure you embark on this journey with the right leadership team.

Another very important thing is that the team should be honest about which old moats are being weakened.

Simply having data is usually not enough.

Integration is becoming increasingly easy to replicate.

When agents can move more easily across systems, workflow and UI advantages become less important . Migration becomes increasingly easier.

Competitors will increasingly target each other's core modules, not just peripheral ones. This means pricing pressure on core businesses is imminent, so prioritize advantages that help you maintain pricing power and customer retention.

This is possible: lessons learned from Broadcom

Before the advent of AI, there was a case study of hardline practices in the public markets: Avago/Broadcom under Hock Tan's leadership. It was a harsh model. It wasn't every founder's cultural blueprint. But it reminded us that aggressive cost control, product simplification, and price achievement were possible. Hardline practices do exist.

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

Key issues

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

Will it achieve revenue growth of over 10 percentage points through brand-new AI products? Or will it achieve a true operating profit margin of over 40%, including SBC?

Investors should ask the same questions more forcefully than they do now.

Where is the AI ​​product engine capable of reshaping the curve? Where is the R&D architecture restructuring centered around small, tokenized, and customer-centric teams? Where are the plans to build a dual human/agent interaction layer? Where is the clear roadmap to achieve a true profit margin of 40-50%+? Where are the plans to reduce equity dilution as a percentage of revenue?

If the answer is some version of "a little of both" or "we are evaluating various options," I expect the market to continue to exert pressure.

Founder: You need to choose a path, and you need to quickly decide who on your team you want to go with. You have the opportunity to create a new entrepreneurial opportunity for your company, your new team, and your investors. Either grow by 10% or earn 40%. Either build the next generation product or build a cash cow. There is no middle ground. Good luck.