The failure modes of Katerra (attempted vertical integration of design, factory, logistics, and GC services simultaneously) and WeWork (capital structure that only worked in a single rate environment) are different problems. Both taught institutional investors the same lesson: a construction platform's story has to hold up when someone stress-tests the unit economics, the regulatory path, and the market demand independently of the deck narrative.
Here are the five filters that tend to stand up to that stress test.
Filter 1: Margin profile at current volume, not projected volume
The common failure mode of construction startups is a projected gross margin that depends on volume the company has never actually hit. The filter is simple: does the unit economics work today, at the current shipment volume, or only after a 10× ramp? If the answer is "only after a 10× ramp," the profile is closer to WeWork's capital math than to a defensible platform.
A stronger signal: positive per-unit gross margin on the first meaningful cohort of units shipped, before factory utilization assumes any growth at all.
Filter 2: Patent moat, stress-tested
Patent count on a deck is the easy number; stress-testing the claims is the work. The useful exercise is:
- Read the numbered claims, not the descriptions.
- Identify the one or two claims that encode the actual product differentiation.
- Ask the founders (and ideally outside counsel) what a well-funded competitor's design-around would look like.
If the design-around path is 18+ months and requires inventing new geometry or new chemistry, the moat is real. If the answer is "redesign an existing part," the moat is a hedge, not a wall.
Lok-N-Blok's portfolio includes patents covering the interlocking block geometry, the post-installation top-plate tensioning system, MEP channel architecture, and the ENVION composite material chemistry. The block geometry and the tensioning system together produce the engineered wind rating — removing or substituting either degrades the stamp.
Filter 3: Code certification, per-jurisdiction
This is the filter generalist VCs under-weight. Construction products live or die by whether the local authority having jurisdiction (AHJ) will stamp the permit. Products that claim national addressability without state-by-state approvals are pricing in optionality they haven't earned yet.
The correct question: where is the product permittable today without a variance or engineering workaround? And what does the 12-month path look like for the next wave of jurisdictions?
Lok-N-Blok's HVHZ-level approval in Florida is the strongest possible residential-wind beachhead in the U.S. — if the wall system holds in HVHZ, it holds in nearly every inland residential code. ICC-ES eligibility means 50-state expansion is a review process, not a re-engineering effort.
Filter 4: LOI pipeline quality, not pipeline size
Every pre-IPO deck shows a headline pipeline number. The number means different things depending on what's in it:
- A "signed LOI" with open out clauses is marketing.
- A "signed LOI with volume and timeline commitments" is a soft intent.
- A "signed LOI with volume, timeline, and exclusivity" is a prospective contract.
- A "signed LOI plus a pilot PO already executed" is a closed sales cycle waiting for a term.
The third and fourth categories are what matter. Ask for the breakdown — what share of the headline pipeline is in each category? A smaller pipeline of high-quality LOIs is worth more than a larger pipeline of category-one intent letters.
Filter 5: Operator credibility + distribution
The signals that tend to show up in platforms that clear Series-A reviews:
Insurance delta as a distribution channel
If the product creates a documented premium reduction in carrier rating tables, that's a sales channel competing systems structurally cannot match. Not a "feature" — a structural competitive moat.
Labor-path-around
The construction skilled-labor shortage is a permanent structural drag on incumbent methods. A system that collapses a multi-trade build into a small certified crew unlocks the portion of the addressable market that cannot be built today because the labor doesn't exist.
Code-official enthusiasm
The most interesting field signal is when code officials actively want the product in their jurisdiction because it solves a problem they already have (hurricane resilience, wildfire code, affordability in high-COLI markets). That's distinct from compliance — it's political pull.
Operator-credible brand advisor
A brand advisor whose career validates the category (Kevin Harrington: original Shark Tank, DRTV pioneer, consumer product distribution) brings rolodex + visual shorthand that the broader category doesn't have. Distinct from a name-for-hire; closer to a distribution partnership.
The moat against "us, but cheaper"
Construction is among the most imitate-able sectors on earth. The worst place to be is "the premium version of a generic product." The best place is "a product that is physically impossible to undercut on total cost of ownership when insurance, service life, and labor are included." Patents + engineering stamp + insurance delta stack into that moat.
The Katerra question, specifically
Every pre-IPO construction deck gets asked some variant of "aren't you just Katerra 2.0?" Katerra's failure mode was trying to own every vertical simultaneously, without de-risking any of them independently. The combined operational complexity crushed the cost curve that made the pitch work.
The lesson: pick the one step that genuinely compounds, and partner out the rest.
Lok-N-Blok manufactures the product and sells through existing builder networks. Manufacturing is partnered (a tooled facility, not an owned plant). Installation goes through certified builder crews, not a captive GC arm. The P&L profile is closer to a materials company than a vertically-integrated real-estate operator — which is the relevant distinction for an LP worried about Katerra 2.0.
How Lok-N-Blok maps to the five filters
Every filter above is answerable with documents, not narrative. The data room behind the NCNDA contains:
- The current financial model with unit-economics at shipped volume.
- The full patent portfolio and outside counsel's review notes.
- The FBC, Miami-Dade, ASTM, and ICC-ES documentation.
- The LOI breakdown by category.
- The cap table, raise terms, and round structure.
If you want to evaluate the raise against the five filters, schedule a founder call. 30 minutes, answered from the model, not the deck.