Technical diligence tests whether a product can serve far more customers without collapsing or destroying margins, by looking at architecture, infrastructure costs and past incidents.
11 min read · Level 3
I do not expect an early product to be ready for scale, and I think it would often be a mistake if it were. What I want to see is a technical team that can name, specifically, what will break and roughly what it will cost to fix, because that shows they understand the system they have built.
Most early products are built to work, not to scale. That is often right; the question is what scaling will require.
Three practical tests:
1. Where does it break? Ask the technical team what would fail first at 10x current usage - the database, a third-party service, a manual process - and what fixing it would take.
2. How do costs scale? Infrastructure cost per customer:
Customers 500 Monthly infrastructure cost 20,000 Cost per customer 40 Revenue per customer per month 150 Infrastructure share of revenue 40 / 150 = about 27%
If cost per customer stays at 40 when customers reach 5,000, infrastructure costs reach 200,000 a month and gross margin remains limited. If well-designed systems bring it down to 15, the margin improves substantially.
3. What has already gone wrong? Outage history, performance incidents and how they were handled show how the system and team behave under stress.
Because a product that must be rebuilt to grow consumes time and money exactly when the company should be selling. And infrastructure costs that grow with every customer cap gross margin, which caps valuation.
Technical founders also often understate these costs, not deceptively but because early usage has never tested them.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Two scaling paths.
Cost per customer at 5,000 customers Monthly cost Share of 150 revenue Unchanged at 40 200,000 about 27% Improved to 15 75,000 10%
The difference is 125,000 a month, or 1,500,000 a year. The investor asks the technical lead what work would reduce the cost, how long it would take, and whether it is in the plan and budget.
Assuming a product that works for current customers will work, and cost the same per customer, at much larger scale.
Identify the likely bottlenecks and the work needed to remove them.
Divide infrastructure and third-party costs by active customers over time.
Look at outages and how quickly and honestly they were handled.
Confirm scaling work is resourced, not assumed.
During diligence on any technology company expecting rapid growth.
Do not penalise very early companies for not having built for scale; judge whether they understand what scaling will require.
Venture investors and industry analysts have written extensively about cloud infrastructure costs significantly reducing gross margins at software companies as they scale, with some large companies publicly discussing efforts to reduce cloud spend or move workloads to improve margins.
Infrastructure cost is a margin question, and margin is a valuation question.
Create a free account to read the rest and apply it to your own investing.