Dallas builds. Houston supplies a different bench.

Why the delivery bench concentrated here

Dallas is where the buildings and the leases are, and the commercial property researchers who track that inventory measure it directly. CBRE put Dallas-Fort Worth colocation inventory at 1,067.3 megawatts at the end of 2025, with just 26.0 megawatts marketed as available and 470.8 megawatts absorbed across the year, in the H2 2025 trends work it published in February 2026. A market that tight is a market in which the scarce legal skill is closing space quickly and cleanly: form leases that survive a hyperscale counterparty’s redlines, service-level credits that an operations team can actually meet, and construction contracts that keep a delivery date.

That inventory also explains the land-use bench, because in the Dallas ring the cities are the gatekeepers. MultiState counted at least 100 local Texas ordinances considered since 1 July 2025, and the Dallas-Fort Worth municipalities supplied a disproportionate share of them: a special-use permit regime in Lewisville adopted in June 2026, a light-industrial classification with a conditional-use permit and a residential buffer in Forney in April 2026, and a dedicated data-center and storage framework in Mesquite in July 2026. Each of those is an application, a staff report, a commission hearing and a council vote, and each of them is a file a head of legal either staffs or outsources.

Sartori’s Dallas file has the same shape. Of the 22 in-house searches closed in Dallas over the trailing three years, on a desk that has worked this market for more than 10 years, 14 were real-estate, construction, procurement or commercial-contracting seats and 5 were corporate, funds or capital-markets seats. Three were energy. The counter-offer incidence on the Dallas in-house book runs at 30 percent and the median gap between offer and acceptance is 12 working days, which tells a general counsel something useful: in Dallas, the competition for a delivery lawyer is real but the close is fast. It is the brief, not the negotiation, that goes wrong here.

The internal spread across those Dallas seats is the more interesting number. The litigation seat topped out at $233,000 and the funds seat at $300,000 at the same class of employer, in the same metro, in the same year: a gap of about 29 percent that has nothing to do with seniority in the conventional sense and everything to do with which part of the capital structure the lawyer touches. The energy seat sat between them at $260,000, which is itself worth noticing, because that is a Dallas employer paying below its funds band for the capability that its own critical path now runs through.

A chief legal officer at a fund-backed campus developer put the consequence plainly to us: the company had budgeted a delivery lawyer and discovered it needed a regulatory one, and by the time it understood that, the energization date had already been committed to an investment committee. Sartori’s Dallas mandate telemetry supports the pattern rather than the anecdote. Of the 22 in-house searches closed in Dallas over the trailing three years, 5 stalled at least once past ninety days, and in 3 of those five the block was on our side of the table: we shortlisted delivery lawyers for a seat whose critical path had already moved to interconnection, and the client was right to send the list back. That is a mapping failure, not a market failure. Our Dallas file records admission, employer and practice label; it does not record who has actually appeared in a commission docket, because no public register does.

A head of legal at a Houston independent power producer described the division of labor to us in terms worth borrowing. On his side of the table, the campus is a counterparty with a credit profile and a curtailment posture; on the operator’s side, the same campus is a building with a delivery date. Both descriptions are complete and neither is sufficient, and the two legal departments that have to transact with each other are staffed from cities that do not interview the same people. He put the practical consequence in one line: the negotiation goes slowly whenever one side has to phone a law firm to understand the other side’s document.

A general counsel at a Dallas-headquartered colocation operator made the complementary point about sequencing. Her department had been sized for volume — a lease book, a construction program, a procurement pipeline — and the change since 2025 was not that the volume fell but that a single low-frequency workstream became the one that decided whether any of the volume mattered. She was explicit that the answer was not to convert the department: it was to add one seat that could hold a regulated position and to stop pretending the utility relationship was a commercial account.

What a realistic timetable looks like

The typical in-house search runs 4 to 7 months from brief to start date. Completion runs at 93 percent across the 22 in-house searches Sartori has closed in Dallas, and at the same rate across the 19 closed in Houston. The variance is not in the sourcing. In Dallas the market is dense enough that a delivery shortlist assembles quickly and the close is fast, at a 12-working-day median between offer and acceptance. In Houston the shortlist is smaller, the candidates are being approached by employers outside the state, and the median gap is 13 working days. What actually consumes the calendar in both cities is the two or three weeks at the start in which a company decides which of its gates it is really hiring against — and the companies that skip that conversation are the ones that run the search twice.

For the service view of either book, the Dallas in-house desk and the Houston in-house desk set out how we brief, map and close these seats. Companies building a legal function from nothing rather than adding to one should start with the first general counsel guide instead, because the scoping question there comes before the geography one.