Since Salesforce was founded in 1999, the SaaS model has completely transformed how enterprise software is delivered worldwide. But in China, after more than a decade, this track is still in its “looks bustling, feels brutal” early stage — of 3,833 operating companies, only a handful clear 100 million yuan in annual revenue, and legions of founders struggle day after day in a red ocean.
Why is it that, selling the same software, some people build Salesforce with billions in annual revenue while others can barely survive? In The SaaS Startup Roadmap (《SaaS创业路线图》), Wu Hao gives a plain but lethal answer: the essence of SaaS is renewal. Not technology, not features, not sales scripts — it’s whether you’ve built a positive loop that makes customers willing to pay year after year. The book systematically lays out the whole startup picture, from market positioning and business models to pitfall-avoidance principles. I’ll share the four dimensions that hit me hardest.
1. Renewal Isn’t a Billing Method — It’s a Mechanism That Guarantees Service Quality
Many people’s understanding of SaaS stops at the surface: “subscribe by year, no servers to buy.” Wu Hao punctures that — the most fundamental difference between SaaS and traditional perpetual-license software isn’t the deployment method; it’s the service quality forced into being by the renewal mechanism.
Selling traditional software licenses is a one-off deal. Once the money lands, the vendor has no incentive to keep serving the customer; only the support team reacts passively when a bug is reported. But SaaS renews annually — if the customer doesn’t renew in year two, the vendor loses that revenue outright. This pressure forces vendors to set up CSM (Customer Success) teams with “customer renewal” as the core KPI.
A good mechanism beats ten thousand rounds of daily management. When a CSM’s bonus is tied to customer renewals, he’ll proactively mobilize product, engineering, even executive resources to solve the customer’s problems — because he only gets paid when the customer is doing well with the product. What’s more, if a customer pays for 3 to 5 years upfront, the vendor actually loses its incentive to serve — which runs contrary to the underlying logic of SaaS.
I paused here for a while: do the many ToB businesses at my company truly have a “customer success” mindset? Or did we just rename after-sales support? If it’s just a rename, then in essence we’re still doing traditional software maintenance — hardly real SaaS.
2. Pick the Right Battlefield: Olive-Shaped Markets Are Standard SaaS’s Home Turf
The most common mistake Chinese SaaS founders make is chasing the budgets of big head enterprises, only to be dragged to death by customization demands. Wu Hao explains this clearly through the contrast between “olive-shaped” and “dumbbell-shaped” markets.
The typical olive-shaped market includes industries like restaurants, education, and fitness studios — a huge number of customers, with mid-size businesses of 50 to 500 people as the core target. These companies have a genuine need for efficiency driven by competition, and they’re unwilling to pay hefty customization fees — a natural fit for standardized SaaS products. At the two ends, micro-businesses have poor renewal rates and large enterprises demand heavy customization; neither suits long-term deep cultivation.
Dumbbell-shaped markets, like petrochemicals, banking, and power supply, have few customers and highly idiosyncratic business processes — you can almost only survive through custom development. Such markets aren’t off-limits, but you need to think it through: do you want to run a SaaS company, or a software outsourcing shop?
This classification was a real eye-opener. In the past, whenever we ran projects, we unconsciously drifted toward big clients with big budgets — a project would take half a year, and by delivery the team was wrecked and margins razor-thin. After reading this book I’m even more convinced: building standard products for the mid-market may look small per deal, but the compounding from scale and repeat purchases far outweighs the temptation of a few big contracts.
3. Business Model Upgrades: From “Selling Shovels” to “Sharing the Gold Mine”
The book ranks SaaS business models in five tiers from low to high; the higher you go, the deeper the binding between vendor and customer, and the greater the commercial value.
The bottom tier is the traditional perpetual license — already doomed to decline. The second tier is annual renewal, the basic SaaS form. The third tier is the prepaid-consumption model, where the customer is charged by actual usage, with higher payment frequency. The fourth tier is very interesting — revenue sharing on sales: the vendor no longer charges a fixed annual fee but embeds itself directly in the customer’s transaction flow, taking, say, 1% of actual transaction volume. That means the better the customer’s business, the more the vendor earns — the two parties go from “tool vendor and buyer” to “operating partners.”
The top tier is the industrial internet model: not being a mere shovel-selling tool vendor, but going deep into the industry’s business, operations, and supply chain, accumulating transaction data, and sharing in the industry’s growth dividend.
This tiering reframed my understanding of the “customer value” we talk about daily. If your product only saves the customer a bit of labor and nudges efficiency, the customer’s cost of replacing you is low. But if you’re deeply embedded in their business processes — or even tied directly to their revenue — then the relationship isn’t about “buy or not buy,” it’s about “can we make money together.”
Don’t be a SaaS that just sells shovels; help the customer find the gold mine and share in it. That line deserves to be pasted on the wall.
4. Pitfall-Avoidance Principles: Calculate the Ceiling Before You Start
The second half of the book devotes a lot of space to the pits SaaS startups must avoid. I’ve summarized the four principles most useful to me.
First, calculate the market ceiling before entering. Wu Hao notes that many Chinese SaaS companies fight their way to #1 in a niche track, only to discover the whole market is worth a few tens of millions — not enough to support a company’s valuation or team. That’s not an execution problem; it’s failing to do the math when picking the track.
Second, refuse vicious price wars. Enterprise needs vary widely; the best strategy is niche differentiation, not racing to cut prices before you’re profitable. Hot money in China’s SaaS industry bred many players who “chase revenue and ignore profit,” but in a fully competitive market, there is no long-term monopolistic windfall.
Third, abandon the fantasy of “monopolize cheap, raise prices later.” That’s a state-owned-enterprise playbook; it doesn’t work in a fully competitive market. Customers you win with low prices have extremely low switching costs — raise prices and they churn.
Fourth, hold to the positive business loop. A quality product solves customer pain points → word of mouth brings new customers → cash flow feeds back into R&D and organizational upgrades → continuous iteration serves customers better. Once this loop breaks — say, insufficient product value leads to poor word of mouth and slow new-customer growth — you enter a vicious cycle that’s hard to escape.
Selected Excerpts
Below are the essences I excerpted from the book, organized by theme, as a memo for further reading.
On the Essence of SaaS
- The essence of SaaS is “renewal.” This is the fundamental difference between SaaS and traditional installed software.
- A SaaS company has an inherent mechanism: it asks customers to renew every year, so it keeps solving customer problems. Internally there’s a CSM (Customer Success Management) team, with a dedicated customer success manager whose KPI is the customer’s “renewal.”
- A good mechanism beats ten thousand rounds of daily management. “Annual renewal” is exactly such a mechanism for guaranteeing service quality.
- If a SaaS company collects 3 or even 5 years of fees at once, the risk to the customer is actually greater than with traditional software — because with no renewal due for 4 years, the CSM won’t prioritize that customer for 4 years.
- In a SaaS company, the customer success department is the most important department. It replaces the old “customer service department.” Customer service is mainly passive response, while a customer success manager must proactively help the customer, to achieve renewal and upsell.
- As for customers who pay but don’t use the product, I consider them not assets but liabilities. From both a financial and a market-reputation standpoint, they’re negative assets.
On Market Positioning and Business Models
- SaaS products best fit “olive-shaped” target markets. The middle of such a market is large, with plenty of “SMBs” (companies of 50–500 people) holding most of the market share.
- Doing custom development for a few head enterprises may be one path for a SaaS vendor to hone a future standardized product, but long-term custom development for big clients is definitely not a SaaS company’s end goal.
- The business model is precisely what determines the height of the ceiling. Many founders assume the future market is huge because they’ve never actually been “#1 in the industry.” Only after years of grinding to #1 in a niche do they realize: they finally sharpened their teeth, but there are hardly any fish left in the pond.
- My yardstick for a SaaS company’s business model is the frequency of generating revenue — that is, the frequency of payment. The higher the payment frequency, the higher the business relevance, the greater the commercial value.
- We’re still just selling shovels — how can we have a stake in the customer’s revenue? We could consider joining the industry’s business layer: help customers find the gold mine, give them methods for mining it, or operate the mining business together — only then do we have a chance to share in the gold that’s dug up.
On Moats
- The moat is switching cost. For an enterprise, choosing a SaaS product involves, beyond purchase cost and procurement-decision cost, usage costs (including process changes, new workflows, and operational training) and long-term maintenance costs.
- The best moat is the network effect. Both tool SaaS and business SaaS products can potentially form network effects.
- There’s another moat called “a more granular industry.” When big players enter, they won’t do just one niche. Already having annual sales of billions of yuan or more, they want at least a new market worth over a billion. If the industry you picked has only a 200-million-yuan opportunity, the giants won’t come.
- All moat-building must first rest on the real, perceivable value the product brings the customer.
On Product Polish and Custom Development
- To-B products can only attack pain points. Consumer demand is fairly emotional; B2B demand is highly rational. Don’t think you can bluff money out of people “now” with something “futuristic”!
- Don’t try to build complete functionality in one go — a new product should win by being “sharp,” not “complete.” That’s also what an MVP is supposed to mean.
- Be clear whether you’re building a product or doing projects. Project-based companies can survive in China and make money, but it’s “headcount money”: this project takes x person-months, at y per person-month, revenue is z=xy.
- Even when doing custom work, only customize within your own boundaries. For customization outside your boundaries, find mature products and just build the interfaces.
- When the right moment appears, convert to a product as early as possible. Custom development capacity has a ceiling, and even adding headcount may not reverse the trend of diminishing marginal returns.
On Marketing and Sales
- The product can truly sell: sell to unfamiliar customers, not just to your old friends’ companies.
- Nothing proves a product’s value better than “on-time payment collection.”
- Rather than spending time placating low performers, spend time seriously studying customers.
- The faster the deal and shorter the sales cycle, the higher the required degree of standardization.
- In a perfectly competitive market, enterprises have no pricing power; supply and demand determine price. “Product differentiation + service differentiation” is the key to profitability.
- In markets with low entry barriers, don’t imagine you can “eliminate” rivals through low prices — even if a rival collapses you can’t raise prices, because once there’s profit, new rivals enter.
- Let’s derive the state of a company forced into a low-price strategy: low price per deal → low personal income for sales reps → can’t recruit excellent sales reps → reps can’t move customers → can only persuade customers with bigger discounts.
- Generally, a “refund clause” is a win-win choice. For an enterprise-grade product the refund ratio is very low — barring major product issues, the proportion of refunding customers, by count and by amount, won’t exceed 1%.
- A good mechanism beats ten thousand rounds of daily management. An extremely dissatisfied customer will demand a refund; refunds will naturally affect every department’s performance metrics, driving improvements across departments.
On Teams and Management
- Certain management laws cannot be violated — for example, the “span of management” cannot exceed 8 people.
- Management ability isn’t a gift; it’s experience.
- Heroes have always risen from the ranks. In innovation and entrepreneurship, willingness matters more than experience. Experience has already settled into the company, while willingness is the engine that drives people to act.
- How do you “screen” for willing people? My suggested sieve is “voting with your own RMB”: core members of a new business team should put up their own money to claim some equity.
- There are no bad employees, only bad systems. If everyone in a company is selfish and harms the company’s interests, maybe the company’s mechanisms (and hence its culture) are what’s broken.
- The greatness of human commercial society: every member, while pursuing personal interest, also realizes social value. There are no bad employees, only bad systems.
- Sales is a bitterly hard profession — just look at how many cold calls get hung up within 30 seconds every day. Who likes being hung up on abruptly or having a cold face thrown at them?
On Customer Success
- In a SaaS company, the customer success department is the most important department. In all seriousness, a SaaS company without a customer success department is a textbook fake SaaS company.
- The first 30 days after purchase are the critical window for whether the customer actually gets the product up and running.
- The first and most important thing in customer success is the customer profile — find the right customers to serve. Many companies, to survive, sign customers outside their profile. That may bring revenue, but it damages the product’s reputation.
- A SaaS company with a renewal rate below 70% isn’t worth much. Put another way, such a company should be valued at traditional-company P/E multiples, not at 10x revenue.
On Financing and Operations
- Compared with traditional software companies, SaaS companies collect software service fees annually, so overall revenue is effectively deferred. That’s why SaaS companies basically all need financing.
- If we express the relationship between “company valuation” and “fundamental value” as a formula: company valuation = fundamental value × coefficient 1 + premium for changes in the future revenue model.
- R&D expense growth must not exceed “new-business growth,” or annual profit stays negative.
- A SaaS company should either persist in building tools whose renewal rate (including upsell from existing customers) exceeds 100%, or provide customers with more value-added value.
- Before scaling sales, first confirm the product truly delivers value to customers. A company’s greatest disaster isn’t a product that won’t sell — if it truly won’t sell, you can still improve the product. The greatest disaster is sales being too strong, selling a product that others can’t sell: the numbers look great on the surface, but the seeds of ruin are buried 6–18 months out.
Seven Consensuses in China’s SaaS Circle
- To-B products shouldn’t be free. When an enterprise procures a product, beyond the purchase cost there are decision costs and training/usage costs — for an enterprise, there is no such thing as a “free product.”
- On choosing custom development: put general needs into the next version in a configurable way; complete an open API platform as early as possible; PaaS requires 3+ years of long-term investment.
- Set up an SDR team under the marketing department. SDR’s collaboration and evaluation relationship with its upstream (marketing) is tighter.
- Value building the sales team’s self-sourcing capability. Marketing is a long-term investment; sales is the final strike — balancing the two is something the CEO must think through clearly.
- The CSM is the owner of the renewal rate. Letting sales reps own renewals creates quite a few problems.
- Tool SaaS in the micro-business market struggles to achieve gross margin. Micro-businesses have low willingness to pay, weak ability to pay, and low demand for management efficiency.
- Avoid multi-year deals where possible. The essence of SaaS is renewal; companies that collect multi-year service fees up front mostly can’t keep up with service afterward, and product improvements easily drift away from market needs.
Final Thoughts
After finishing the book, my deepest takeaway: the biggest pain point of China’s SaaS industry isn’t insufficiently advanced technology — it’s that the enterprise efficiency mindset hasn’t awakened yet. Starting a company in a market with $2,880 of SaaS spend per capita versus a market where SMBs generally haven’t realized IT systems can cut costs and boost efficiency — the difficulty isn’t even on the same order of magnitude.
But that’s precisely the opportunity. As Wu Hao says, SaaS + AI + IoT are the core tools for restructuring every industry. In soil where demand hasn’t been fully educated, whoever first builds a product that truly solves pain points, establishes a positive renewal-based loop, and picks the right olive-shaped market to cultivate deeply — that’s whoever may seize the initiative when the industry takes off.
For those of us also groping our way through the ToB world, this book’s greatest value isn’t some magical startup secret. It’s helping you think through the three most basic questions — what to sell, whom to sell to, and how to collect payment — before you start. Thinking it through before acting is itself the biggest pitfall avoided.
评论
0 条评论