Latest edition
Enterprise Procurement Pulse — Q3 2026
32 published observations · 43 observations reviewed · 3 sources · compiled 9 September 2026, verified 10 September 2026
Every figure below carries its source, the date it was published, and the equipment class it describes. Figures are never averaged across source types or across equipment classes. Where a source does not state the class behind a number, that is noted rather than inferred.
Source types: Research source. Research · Practitioner source. Practitioner · Vendor source. Vendor · Government source. Government. What these mean
Correction, 18 September 2026. We removed a Wood Mackenzie range for large power transformers (80–210 weeks) from the current comparison: its source is a 2 April 2024 article based on Q4 2023 data. Source labels now show when each source was published; earlier labels for JLL, SourceBlue and SAVRN showed an edition name or year instead.
Correction, 29 September 2026. We removed SAVRN’s figure of 26–32 weeks for standard medium-voltage switchgear configurations. SAVRN’s guide does not identify an original source or its own measurement for that figure, and we could not trace it to an original publisher.
About the JLL figures. JLL values are derived from the Americas bars in its January 2026 report (PDF), page 24, rounded to the nearest week. The chart does not label individual bar values or specify the measurement period or sample.
About the counts. The tables below hold 33 table rows but 32 published observations. JLL’s single switchgear benchmark states no voltage class, so it appears under both medium- and low-voltage switchgear, labelled there as one shared benchmark rather than two measurements.
Transformers
| Class | Lead time | Source |
|---|---|---|
| Dry type, 0–2 MVA | 20–32 wks | Practitioner source. Terrapin · Jun 2026 |
| Pad-mount, 0–5 MVA | 40–65 wks | Practitioner source. Terrapin · Jun 2026 |
| Substation, 5–25 MVA | 65–95 wks | Practitioner source. Terrapin · Jun 2026 |
| Substation, 25–50 MVA | 85–110 wks | Practitioner source. Terrapin · Jun 2026 |
| Generator step-up, 50–150 MVA | 100–150+ wks | Practitioner source. Terrapin · Jun 2026 |
| Category benchmark | ≈43 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
DCU note. JLL’s Americas transformer benchmark (about 43 weeks) does not identify the transformer class behind the figure. The banded classes above range from 20 weeks to more than 150, depending on type and capacity. Which row applies depends on what is actually being specified.
Generators
| Class | Lead time | Source |
|---|---|---|
| 500–1,000 kW diesel | 32–48 wks | Practitioner source. Terrapin · Jun 2026 |
| 1,500–2,000 kW diesel | 50–66 wks | Practitioner source. Terrapin · Jun 2026 |
| 2,500–3,500 kW diesel | 62–82 wks | Practitioner source. Terrapin · Jun 2026 |
| 3,500–4,500 kW diesel | 90–110 wks | Practitioner source. Terrapin · Jun 2026 |
| 1,000–3,000 kW natural gas | 54–78 wks | Practitioner source. Terrapin · Jun 2026 |
| Size-dependent range | 28–66 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| Category benchmark | ≈51 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
DCU note. Terrapin reports 32–48 weeks for 500–1,000 kW diesel generators; JLL’s Americas generator benchmark reads at about 51 weeks. JLL does not state the size class behind its figure, so the two measure different scopes. The difference does not show which figure fits a particular generator specification.
Also recorded, not published above: Terrapin’s summary figure of 50–78 weeks for 1,500–3,000 kW diesel, which compresses several of their detailed bands.
Medium-voltage switchgear
| Class | Lead time | Source |
|---|---|---|
| Pad-mount loop switch | 32–50 wks | Practitioner source. Terrapin · Jun 2026 |
| 5 kV / 15 kV metal-enclosed | 52–72 wks | Practitioner source. Terrapin · Jun 2026 |
| 15 kV / 27 kV metal-clad | 60–80 wks | Practitioner source. Terrapin · Jun 2026 |
| 38 kV | 78–104 wks | Practitioner source. Terrapin · Jun 2026 |
| 38 kV / 69 kV, customized | 96–120 wks | Practitioner source. Terrapin · Jun 2026 (FAQ) |
| Size-dependent range | 28–48 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| All switchgear, voltage class not stated (one shared benchmark, also shown under low voltage) | ≈43 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
DCU note. The reported ranges vary substantially by voltage class and construction — a pad-mount loop switch and 38 kV gear differ by more than seventy weeks. The customized 38/69 kV figure comes from the same source’s FAQ rather than its detailed table and may describe a different population; both are shown rather than reconciled.
Low-voltage switchgear
| Class | Lead time | Source |
|---|---|---|
| Size-dependent range | 26–58 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| All switchgear, voltage class not stated (one shared benchmark, also shown under medium voltage) | ≈43 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
Limited coverage in the current dataset. No banded figures by amperage or configuration. JLL reports a single switchgear figure with no voltage class; it is the same benchmark shown under medium-voltage switchgear, not a separate low-voltage measurement.
UPS
| Class | Lead time | Source |
|---|---|---|
| 500–1,500 kVA | 30–48 wks | Practitioner source. Terrapin · Jun 2026 |
| Size-dependent range | 16–42 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| Category benchmark | ≈35 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
DCU note. JLL’s Americas UPS benchmark (about 35 weeks) falls inside Terrapin’s 30–48 week band for 500–1,500 kVA. For generators, JLL’s benchmark falls above Terrapin’s 500–1,000 kW range. A broad benchmark may or may not fall within the range relevant to a given project, and the category figure alone does not indicate which.
Cooling
| Class | Lead time | Source |
|---|---|---|
| CRAHU and FWU | ≈11 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
| Cooling tower | ≈14 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
| RDHx and DTC | ≈19 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
| Chiller, air cooled | 3–49 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| Chiller, water cooled | 12–60 wks | Practitioner source. SourceBlue · Q2 2026 report (Aug 2026) |
| Chiller, category benchmark | ≈35 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
Category labels reproduced as JLL published them. JLL does not define CRAHU, FWU or DTC, or state whether the RDHx and DTC figure includes coolant distribution units.
DCU note. The air-cooled chiller range spans 3 to 49 weeks, which indicates the band covers a wide span of equipment sizes. The current verified dataset has limited tonnage-specific chiller coverage.
Distribution and switching
| Class | Lead time | Source |
|---|---|---|
| ATS | ≈35 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
| PDU / STS | ≈41 wks | Research source. JLL · Jan 2026 · Americas (chart reading) |
Single-source coverage. No banded figures in the current dataset.
What this quarter’s data shows
Equipment class can produce more variation within one source than is visible between broad category benchmarks. Terrapin’s transformer bands run from 20 weeks to more than 150, all from one source in one quarter under one methodology. Their MV switchgear bands span more than seventy weeks.
A category benchmark may or may not fall inside the relevant band. JLL’s UPS figure lands inside Terrapin’s 500–1,500 kVA range. Their generator figure sits above Terrapin’s 500–1,000 kW range. Nothing in the category figure indicates which case applies.
Summary ranges are not interchangeable with detailed bands. Terrapin publishes both, and the summary compresses several detailed bands into one range. A figure quoted from a summary table may not match the banded figure for the same equipment.
Coverage is uneven. Transformers, generators and MV switchgear have class-banded figures from a practitioner source. Chillers have practitioner ranges but no tonnage or equipment-class bands. LV switchgear, ATS and PDU have no banded figures at all in the current dataset, which makes those numbers harder to apply to a specific project.
Methodology and limits
These figures are compiled from published sources. DCU does not quote equipment, does not have transactional visibility of its own, and does not present these ranges as prices or as commitments from any manufacturer.
Lead times vary by manufacturer, configuration, quantity, voltage class and order timing. Terrapin states their figures derive from active 2026 manufacturer slot reservations and TCG project procurement data. JLL’s figures here are read from an unlabelled chart in its January 2026 report (see the note at the top of this page). SourceBlue operates in procurement and supply chain execution.
Where a figure could not be traced to its original publisher, it is omitted from this page. Full approach: Methodology.
Next update: Q4 2026.