Energy / Utilities · Cold Call
The Energy / Utilities Cold Call Script: Thirty Seconds with a Head of Asset Management
The person you are about to ring has a transformer fleet where a big share of the pole-mounts went in during the same build-out wave in the seventies and eighties, which means they don't fail one at a time — they fail in clusters on day three of a heatwave when the overnight minimum never drops below 26 and nothing gets thermal recovery. They know the age curve. What they don't know is which of the four thousand units on the worst feeders is cooking right now, because the only condition data they have is whatever a linesman wrote on a visual inspection sheet two years ago. If you can say that out loud in one breath, you have earned thirty seconds. If you open with your Series B, you have not.
This is a cold call, so the job is not discovery and it is definitely not a demo. It's relevance density. Every second spent on your company is a second not spent on something true about their world — SAIDI minutes that feed STPIS, GSL payments to worst-served customers on a 90km rural spur, vegetation spend per line kilometre that climbs every year while the regulator asks whether it's risk-targeted or just cycle-driven. These buyers are unusually good at spotting someone who has read a brochure about utilities versus someone who has sat in a control room during a storm event.
Expect friction that is specific to this industry, not generic stalling. "Nothing goes on an energised asset without type approval." "We're three years into a five-year determination and the allowance is locked." "We bought a condition monitoring platform in 2021 and nobody logs into it." None of those are brush-offs — they're accurate descriptions of how the business works. Agree with the process, then shrink the first step until it fits inside the gate they haven't closed. The ask on this call is a twenty-minute calendar hold with a named engineer, not a good chat.
The cold call script
Say it in your own words. The structure is the part that matters.
- 1
Before you dial — 90 seconds of prep, out loud
Answer these four out loud before the phone rings. If you can't, don't dial yet. 1. WHO: Exact title and how long in seat. "Manager Network Performance & Reliability, eighteen months in the role, came up through the control room." A Head of Asset Management two years from a reset has a completely different call from a Chief Engineer six weeks into the job. 2. TRIGGER: A determination submission being drafted. A major event review published. A regulator's final decision on their reliability program. A batch of GSL payments after a storm. A job ad for a Principal Distribution Engineer, condition monitoring. If there's no trigger, the fleet is the trigger — every distribution business with a 1970s–80s build-out wave has the same age curve. 3. THE ONE PROBLEM: Write one sentence. Not three. For a Head of Asset Management it's cluster failure of pole-mounts in heat. For a Vegetation Management Program Manager it's defending spend per line kilometre. For Regulatory Affairs it's the evidence base behind the next submission. 4. THE ASK: "Twenty minutes, Thursday 8:15 or Friday 2." Specific length, two specific slots. Have the follow-up email drafted in a window before you dial so you can send it while they're still on the line.
- 2
The opener (0–8 seconds) — pass the 'is this a robot' gate
"Morning David — it's Priya at Kestrel. We haven't spoken, this is a cold call. Can I have thirty seconds to tell you why I rang, and then you can tell me to get lost?" Then STOP. Two full seconds of silence. Do not fill it. Tonality: slower than feels natural — cold call nerves speed you up about a third. "Can I have thirty seconds" ends on a DOWNWARD inflection. Up sounds like begging. Do not say: "How are you today?", "Did I catch you at a bad time?", "I'll be brief." Pattern-interrupt variant: "David — Priya, Kestrel. You don't know me. Bad time?" (Most reply "depends what it's about," which is permission.)
- 3
The reason you rang — Head of Asset Management / Chief Engineer
"We work with distribution businesses on the ageing distribution transformer fleet. The thing that keeps coming up with asset managers is that the pole-mounts don't fail one at a time — they went in during the same build-out waves so they age together, and you get a cluster on day three or four of a heatwave when the overnight temps don't drop and nothing gets any thermal recovery. Everyone knows the age curve. Nobody can tell you which specific units are cooking, because the condition data is whatever a linesman noted on the last visual. So it's run-to-failure with a small proactive program, and every one of those is an unplanned outage, an emergency truck roll and sometimes a pole-top fire report. Is that anywhere near your world, or have you got the unplanned failure rate per thousand units heading the right way already?" Note the exit at the end. It makes the honest answer safe. Do not name a product. Do not say "platform."
- 4
Variant — Manager Network Performance & Reliability
"We work with network performance teams on unplanned SAIDI. What we hear constantly is that the minutes are dominated by things below the recloser that you can't see in real time — control room's got SCADA at the zone sub and customer calls, and everything past that is inferred, so crews get sent to patrol the feeder to find the fault while the ETRs slip and the call centre gets hammered. Then there's a major event review afterwards asking whether it was predictable. And the number itself isn't internal — it's feeding STPIS at the end of the period. Is unplanned SAIDI on rural feeders the thing you're defending at the moment, or is it worst-served customers and GSL?"
- 5
Variant — Vegetation Management Program Manager
"We work with vegetation program managers on the spend-per-line-kilometre question. The pattern is the same everywhere — cycle-based trimming across thousands of line kilometres, contractor rates up, wetter growing seasons, tighter clearance standards, so the number goes up every year. Cut it and unplanned SAIDI on the rural feeders goes up. Grow it and the regulator asks whether the spend is risk-targeted or just cycle-driven. What program managers tell us they can't answer is which spans actually justified the truck and which got trimmed because the cycle said so. Is that a live argument for you this year, or have you already got the risk-targeting evidence sorted?"
- 6
Two questions maximum — then close
You are not running discovery. Two narrow, factual questions, then ask for the meeting. Pick two: - "How are you doing it today — is that a CBRM model or is it age-and-inspection?" - "Roughly how many distribution transformers in the fleet, and how many are past forty?" - "Is that in the current reliability program or is it living with it until the next reset?" - "Who owns the asset criticality ranking — you or the planning group?" - "When you get a cluster in a heatwave, is that showing up in unplanned SAIDI or is it getting excluded as a major event?" Do NOT ask: "what are your top three priorities," "walk me through your process," "what would solving this be worth." Those belong in the meeting you haven't booked yet. Listen for the admission — "yeah, it's a mess," "we've been meaning to look at that," "we lost a hundred and forty units in January." The moment you hear it, stop asking and start closing.
- 7
The ask
"Here's what I'd suggest. Twenty minutes — not a demo, no deck. I'll walk you through what we found when we ran the same exercise on another distribution business's historical SCADA, outage and asset records: no hardware, nothing on an energised asset, nothing touching the OT network. You tell me whether it's worth a second conversation. I've got Thursday 8:15 before your day starts, or Friday at 2. Which is less bad?" Elements that lift the take rate: name the length (twenty, not thirty), say what it is not, give two slots, state the exit. If they say yes: "Sending the invite now while we're talking — can you tell me it's landed? And is there anyone else you'd want in the room, your Principal Distribution Engineer or whoever owns the works program?"
- 8
The soft no and the recycle
"No problem at all. Last thing and I'll let you go — is this a timing thing or a not-a-priority thing?" [Listen. Then:] "Got it. When does the next determination submission get drafted — is that a this-year exercise or is it still eighteen months out? … And who owns the reliability program business case, is that you or Regulatory Affairs? … Perfect. I'll come back to you in March. I'll send one email so you've got my name and you can ignore me until then." A call that produces a dated reason to ring back and the name of the person who owns the business case is not a failure. A call that produces "send me an email" and nothing else is.
- 9
Gatekeeper / EA screen
"It's about their distribution transformer failure rate — Priya at Kestrel, she won't know me, it's a cold call. If she's not the right person is it the Chief Engineer or the network planning manager I should be asking for?" Short, calm, specific. Never claim a prior relationship, never say "it's personal," never be cute. The EA in a utility has heard four vendors this week and will remember the one who lied.
- 10
Voicemail — under 20 seconds, no ask
"David, Priya at Kestrel — you don't know me. Calling about distribution transformer cluster failures in heat events; it's a thing we see in every fleet with a seventies build-out wave. I'll try you again Thursday morning. 0412 555 148." The voicemail's job is name recognition for the next attempt and the email. No ask, no value prop, no callback request.
- 11
The follow-up email — send it while they're on the phone
Subject: Our call just now — transformer cluster failures "David — as promised. Three lines. What we do: retrospective analysis on a distribution business's own historical SCADA, loading, outage and asset data to rank which distribution transformers are most likely to fail next. No hardware, nothing on an energised asset, nothing on the OT network, so it sits outside type approval and security architecture review. What you said: [use their exact words — 'we lost a batch on the Kalinda feeder in January and none of them were on the replacement list']. Thursday 8:15, invite sent. If anything changes I'll work around you." Use their words, not yours. Log those words — they go into the discovery call.
How the call actually sounds
Prospect on the left, the rep on the right.
Rep
Morning David — Priya at Kestrel. We haven't spoken, this is a cold call. Can I have thirty seconds to tell you why I rang, and you can tell me to get lost?
Buyer
I'm walking into a pre-start in about four minutes, so thirty is what you've got.
Rep
Fair enough. We work with distribution businesses on the ageing distribution transformer fleet. The thing asset managers keep telling us is that the pole-mounts don't fail one at a time — they went in during the same build-out waves so they age together, and you get a cluster on day three or four of a heat event when the overnight minimum doesn't drop and nothing gets thermal recovery. Everyone's got the age curve. Nobody can say which specific units are cooking, because the condition data is whatever a linesman noted on the last visual inspection. Is that anywhere near your world, or is the unplanned failure rate per thousand units already heading the right way?
Buyer
Look, I'll be blunt with you. A pole-mount is a four-thousand-dollar asset. We run them to failure on purpose. It's not economic to monitor them individually and I've told three vendors that this year.
Rep
You're right on the asset value and I'm not going to argue that. The cost isn't the transformer, though — it's the outage minutes and the callout. When one lets go at nine at night on a rural spur, what does that actually cost you by the time you've got the crew out, the customers restored and the minutes into unplanned SAIDI?
Buyer
More than four grand, obviously. But we're not putting sensors on forty thousand pole-tops. Anything that goes on an energised asset needs type approval and a security architecture review if it talks to anything, and that's a two-year process before it gets near a live pole. That's not me being difficult, that's just how it works.
Rep
Agreed, and I wouldn't try to shortcut it. Which is why the first step I'd propose has none of that in it. No device, no hardware, nothing touching the OT network. It's a retrospective analysis on data you already have — historical loading, SCADA at the zone sub, fault and switching records, outage history, whatever's in the asset register. We rank the fleet and you go and check the ranking against what actually failed last summer. Type approval doesn't apply because there's nothing to approve.
Buyer
Our register's a shambles, though. Half the rural fleet has the wrong install date and we've got no nameplate data at all for anything we inherited in the amalgamation. You'd be modelling on garbage.
Rep
That's normal, and honestly it's usually the reason to start rather than wait. Install date matters less than you'd think — loading, temperature exposure, fault history and switching records carry more signal than a clean register. And I'd rather scope it honestly: we'd model the portion of the fleet where the data's usable, and the exercise tells you exactly which records to fix first. What proportion of the fleet would you guess has decent loading data behind it?
Buyer
Maybe half. Sixty percent on the urban feeders. But even if this worked, we're three years into a five-year determination. There's no allowance for it. Nothing new gets funded mid-period unless it comes out of someone else's project.
Rep
Understood, and I'd rather position for the reset than pretend the cycle doesn't exist. Two things. One — when's the next submission being drafted, and who owns the reliability program business case? Because that's the real timeline and a submission is a lot stronger with two years of your own fleet data behind it than a vendor's claim. Two — in the meantime, if it defers even a handful of replacements or takes truck rolls out of unplanned faults, does that sit inside the maintenance opex rather than needing new capex?
Buyer
Submission drafting starts late next year. And I'll tell you the other thing — I've got engineers who've run this network for thirty years. They'll tell you which feeders are the problem without a model.
Rep
They're almost certainly right about the feeders. That's not the gap. The gap is which of the four thousand units on those feeders goes first, and roughly when. So here's the honest version of the offer: run it blind, put the ranked list next to what your engineers would have written, and let them be the judges. If it just confirms what they already know, that's a cheap validation and you've lost nothing. If it flags fifty units nobody was watching, that's the conversation.
Buyer
Alright. That's not the worst pitch I've heard this quarter. Send me something and I'll take it to the asset committee.
Rep
Happy to, but a PDF gets forwarded once and dies — you know that better than me. What does a paper actually need to survive your asset committee: cost-benefit, risk quantification, alignment to the reliability program?
Buyer
All three, plus it has to say something about safety or the chief engineer won't back it.
Rep
Then let me build it with you rather than at you. Twenty minutes — no demo, no deck. I'll bring what the ranked-list exercise looked like at another distribution business, and we'll write the safety line together, because the real one is that a failure found in advance is planned switching in daylight with a permit-to-work, not a night callout to a live fault next to a burning pole-top. Thursday 8:15 before your pre-start, or Friday at 2?
Buyer
Thursday. Early. And bring the safety framing, that's the bit that'll matter.
Rep
Done. Sending the invite now while we're talking — tell me when it lands. And would you want your Principal Distribution Engineer in the room, or do you want to hear it first?
Objections you will hear
What they say, and what you say back.
| Objection | How to answer it |
|---|---|
| “Anything grid-adjacent needs two years of pilots and compliance review before it goes near an energised asset.” | Agree with the process rather than arguing it, then shrink the first step until it fits outside the gate. "Completely fair, and I wouldn't try to shortcut type approval. What's the actual gate sequence for you — type approval, OT security architecture review, chief engineer sign-off? Because the first step I'd propose triggers none of them: a retrospective analysis on your own historical SCADA, outage and asset data. No hardware, no device on a pole, no connection to the OT network. Your engineering team gets something to evaluate on their terms while the approvals run in parallel, and inside a few weeks you know whether the model finds anything real on your fleet." |
| “We plan capex in five-year regulatory cycles and this period is locked. There's no allowance for this.” | Don't fight the cycle, position for it — and test the opex line while you're there. "Understood, and I'd rather aim at the reset than pretend the allowance is flexible. Two questions: when does the next submission get drafted, and who owns the reliability program business case? That's the real timeline for me. Separately — if this defers even a handful of transformer replacements or takes truck rolls out of unplanned faults, does it sit inside maintenance opex rather than needing new capex? And either way, the submission is much stronger with two years of your own fleet data in it than with a vendor's claim, so building the evidence base now costs you nothing at reset time." |
| “We bought a condition monitoring platform three years ago and nobody uses it.” | Go straight at the post-mortem rather than defending your own product. "That's common enough that I'd rather ask what killed it. Usually it's one of three things — alerts nobody trusted, a separate login the control room never opened, or output that never made it into the asset register or the works program. Which was it for you? Because if we can't surface inside your ADMS and drop straight into the works program the planners already run, we'll die exactly the same way and I'd rather know that now. Who owns that system? I'd rather talk to them than around them." |
| “Our asset data is a mess — half the transformer records have the wrong install date and there's no nameplate data for the rural fleet.” | Scope down honestly. Utilities respect that more than a claim that data doesn't matter. "That's the normal starting point and it's usually the reason to start rather than wait. Install date matters less than people assume — loading, temperature exposure, fault history and switching records carry more signal than a clean register. Realistically we'd model the portion of the fleet where the data's usable, maybe sixty percent, and one of the outputs is a ranked list of exactly which records are worth fixing first. I'd rather tell you that upfront than promise full coverage and hand you rubbish." |
| “How does this improve safety? If it doesn't, it's not a priority this year.” | Make it concrete, not implied. "Then let me put it in the terms the chief engineer would use. A failure you find in advance becomes planned switching in daylight with an access authority and a full permit-to-work. A failure you don't find becomes a night callout to an energised fault, a crew working near a burning pole-top, and kilometres driven under fatigue rules. Every unplanned emergency response you remove is one fewer live-line job and one fewer callout in the TRIFR exposure. Reducing unplanned reactive work is a safety outcome, and it's one the board understands without translation." |
| “We've got engineers who've run this network for thirty years. They know which feeders are the problem.” | Never imply they don't — recruit them as judges. "They're probably right about the feeders. The gap isn't which feeder, it's which of the four thousand units on that feeder goes first and roughly when. So run it blind: we produce a ranked list, your engineers write theirs, and you compare. If it just confirms what they already know, that's the cheapest validation you'll get all year. If it flags units nobody was watching, that's the conversation worth having. Either way they're the ones assessing it, not the ones being replaced by it." |
| “Send me something and I'll take it to the asset committee.” | Trade the PDF for information and a date. "Happy to — but you know as well as I do a deck gets forwarded once and dies. What does a paper need to contain to survive your asset committee? Cost-benefit, risk quantification, alignment to the reliability program? Let me build that with you instead of sending something generic. When does the committee next sit? Give me fifteen minutes the week before to prep you properly, and then I'll want to know what they asked even if I'm not in the room." |
| “Not interested. (fired in the first three seconds)” | One calm reframe, slower and lower, and only one. "Totally fair, you don't know what I do yet. If I'm wrong I'll hang up — is cluster failure of the pole-mount fleet in heat events something your team's dealing with, or genuinely not a thing for you?" If the second no comes, take it cleanly and ask for the recycle: "No worries. Who owns unplanned SAIDI on the rural feeders — is that you or the network performance manager?" |
Questions reps ask about this call
- Who should I actually be cold calling inside a distribution business?
Start with the person who owns the number your product moves. For transformer failure and fleet risk, that's the Head of Asset Management or the Chief Engineer / Principal Distribution Engineer. For outage minutes and STPIS exposure, it's the Manager Network Performance & Reliability. For spend-per-line-kilometre defence, it's the Vegetation Management Program Manager. For anything that needs to land in the next submission, the Manager Regulatory Affairs is a route in rather than a buyer. Avoid opening on the GM Operations / Field Services unless your pitch is genuinely about truck rolls and crew time — they'll redirect you and you'll have burned the first impression.
- Is cold calling even viable into utilities, given how long procurement takes?
The sales cycle is long, but that's an argument for calling earlier, not for not calling. The gating factor isn't procurement, it's the regulatory determination period — so the highest-value outcome of a cold call is often not a meeting about buying, it's finding out when the next submission is being drafted and who owns the reliability program business case. Two years of an engineering team's familiarity with your model before a reset is worth far more than a fast pilot that dies at the OT security review. Log the reset date as your real close date.
- How do I handle 'this needs type approval' without losing the call?
Agree with it immediately and then shrink the first step so none of the gates apply. Ask what their gate sequence actually is — type approval, security architecture review and network segregation assessment, chief engineer sign-off — and propose a retrospective data exercise instead: their own historical SCADA, loading, outage and switching records, no hardware, no OT connection. It gives the engineering team something to evaluate on their terms while approvals run in parallel. Arguing that your device is safe is the fastest way to be classed as a vendor who doesn't understand energised assets.
- What metrics should I put in the opener?
Two, at most, and only ones the person you've dialled is personally accountable for. Unplanned distribution transformer failure rate per 1,000 units and unplanned SAIDI work for an asset manager. STPIS incentive dollars and GSL payments to worst-served customers work for a network performance manager. Vegetation spend per line kilometre year on year works for a program manager. Use them as things they're measured on, never as claims about what your product delivers — a made-up percentage is the fastest way to be dismissed by an engineer who knows the real range.
- What's a good outcome if they won't book a meeting?
A dated reason to call back plus one piece of intelligence. Specifically: when the next determination submission is drafted, who owns the reliability program business case, whether their existing condition monitoring platform is actually being used, and what proportion of the fleet has usable loading data. "Send me an email" with nothing attached to it is a polite hang-up and produces no pipeline. Trade the email for two facts or a callback date.
- How do I practise this without burning real prospects?
Run the hard versions out loud before you dial. The three that break reps in this industry are the four-thousand-dollar-asset economics pushback, "we're mid-period and the allowance is locked," and the thirty-year engineer defence — all of which are delivered flatly and reasonably rather than aggressively, which is what makes them hard to stay conversational through. Roleplay them in DrillCall as a Chief Engineer who is polite, correct and unimpressed, until your response to resistance is to slow down and drop your tone rather than talk faster.