Envio Growth Engineer — Video Script

Format: Loom-style screen-share + talking-head, ~4 minutes Audience: the Envio team, sent privately alongside the four-document package Voice: operator-confident — crisp, numbers-forward, peer tone Hook: "Here's the math behind why this role pays for itself." Why this exists: to compress the strategy + math memos into one watchable artifact the team can absorb in 4 minutes, and to demonstrate I can actually produce the kind of customer-voice content I'm proposing as Play 3.

Production notes (read before recording)

• Tools: Loom or any screen-recorder with webcam overlay. Webcam in top-right corner, ~20% screen size. Don't hide your face — the talking head is what makes this not feel like a slideshow.
• Audio: wired headset mic or USB condenser. Phone mic is acceptable if it's quiet. Do not record with laptop built-in mic — it's the single biggest tell of "I rushed this."
• Background: clean wall, neutral light, no clutter. Daytime natural light beats any ring light.
• Wardrobe: plain solid-colour shirt or hoodie. No logo'd merch (avoids looking like an outsider performing belonging).
• Take strategy: record in 3 segments — Cold Open / Math / Close — then stitch. Don't try to nail 4 minutes in one take. If a segment goes off, restart only that segment.
• Pace target: ~150 words per minute. Roughly 600 words total for a 4-minute video. The script below is paced for that.
• Total prep time: 90 minutes including 2–3 takes per segment. Don't over-rehearse; on the 4th take you start sounding scripted.

What's on screen, when

Segment 1 — Cold Open (0:00–0:25, ~60 words)

Visual: Full-frame webcam. Black title slide visible behind/around you with text: "The Math Behind a Growth Role at Envio." No music. Just you, talking.

Talk track (deliver crisply, eye contact, no smile yet):

"Hi, I'm Kaustubh. You've already read the strategy memo and the math memo I sent over. This 4-minute video isn't a re-pitch of those documents — it's the version where I walk you through the math live, on screen, so you can see the model the way I'd present it in our discussion.

If the documents are the homework, this is the whiteboard."

Direction note. Don't introduce yourself for 30 seconds. The team knows who you are. Skip the niceties — the first sentence has to earn the next 4 minutes of their attention.

Segment 2 — Mirror Protocol as Proof (0:25–0:55, ~75 words)

Visual: Webcam moves to top-right. Main screen: live Mirror Protocol dashboard. Feed updating with real trades. Scroll slowly so the live tx hashes are visible.

Talk track:

"Quick grounding before the math. This is Mirror Protocol — live on Sepolia, indexed entirely on Envio's hosted service. Seven entities. Multi-event handlers. Real Uniswap V2 swaps. Every row in this feed is a real on-chain trade my executor bot fired in the last few hours.

The reason I'm showing you this first is that every number in the next two minutes is grounded in lived experience. I've already executed the customer's side of the role I'm applying for."

Direction note. Click into one of the tx hashes briefly — let them see the Sepolia Etherscan link open. Half a second is enough. The visual proof is what makes the next section land.

Segment 3 — The Activation Asymmetry (0:55–1:25, ~75 words)

Visual: Switch screen to the published Medium case study, scroll to the section that talks about the 4.2-second → 3-5ms latency improvement.

Talk track:

"Here's the part of the journey that matters for the role. My first version of the bot used direct RPC calls — 21 calls per cycle, 4.2 seconds of decision latency, dashboard felt dead. Switching to Envio brought that to 3 to 5 milliseconds. About a thousand-fold improvement.

That experience is the foundation of how I think about activation. I lived the activation gap. Most candidates argue the gap exists. I have a tx hash that proves it."

Direction note. The line "Most candidates argue the gap exists. I have a tx hash that proves it" is the strongest line in the script. Land it. Pause for half a second after.

Segment 4 — The Math, On Screen (1:25–3:10, ~270 words — the longest segment)

Visual: Switch screen to ENVIO_REVENUE_MATH.md, scrolled to Section 6 — "Plug Your Numbers In." Webcam stays top-right. As you talk through each lever, scroll slowly down the formula. Don't rush — let the team see the math.

Talk track:

"Here's the model. Five revenue levers. Each one has a defensible formula with placeholder inputs in italics — and these are placeholders, deliberately. You replace them with Envio's real numbers on day one.

[Scroll to ACTIVATION LEVER block]

Activation, first. If your monthly free-tier signup volume is 300, and the industry-standard 35% never deploy a working indexer, that's 105 lost users every month. The 'First Indexer in 30 Min' pillar artefact recovers a quarter of those. At 5% paid conversion over 12 months, that's 16 net-new paying customers a year. At a placeholder 4,000-dollar mid-market ACV, that's roughly 64,000 in incremental year-one ARR — from one Loom video, one repo, one next-step doc. Roughly a working week of effort.

[Scroll to MIGRATION LEVER]

Migration. Five completed migrations in year one. Each costs about five working days of my time. At an 18,000-dollar migration-tier ACV, that's 90,000 incremental year-one ARR — and four sales-asset case studies that close the next two prospects each.

[Scroll to CASE-STUDY LEVER]

Case studies. The play with the longest half-life. Four shipped in year one. Each generates roughly five inbound enterprise inquiries within 60 days. At 20% close rate and 50,000 enterprise ACV, the math conservatively recognises about 80,000 in year one — with the bigger payoff back-loaded into year two and three.

[Scroll to TOTAL]

Total at placeholder values: about 286,000 of incremental year-one ARR. The defensible band — half-on-everything pessimism to two-times-on-everything optimism — is roughly 140,000 to 570,000. Against a typical Growth Engineer total cost of 120 to 160 thousand, the role pays for itself across the entire band."

Direction note. This is the hardest segment to deliver crisply. Don't read the numbers — speak them like you're explaining to a colleague. Scroll while talking, not before each block. If you fluff a number, restart that line — but don't restart the whole segment. The math memo is what they're seeing; you're just narrating.

Segment 5 — Why I'm Showing You This Way (3:10–3:40, ~75 words)

Visual: Switch to a static slide listing the four-document package:

ENVIO_REVENUE_MODEL.md      — Strategy
ENVIO_REVENUE_MATH.md       — The math (this video)
ENVIO_GROWTH_PLAN.md        — Operations
ENVIO_DECK_OUTLINE.md       — Meeting deck
ENVIO_FIRST_24_HOURS.md     — Day one

Talk track:

"You're looking at the math because I deliberately chose numbers that are falsifiable, not impressive. Every input you just saw is either a real benchmark, a real Mirror Protocol metric, or an explicit placeholder labelled as such. None of them are confident point-projections at your business. That's the whole point.

The version of this model that matters is the one we co-edit in week one with your real signup volume, your real conversion rates, your real ACV bands."

Direction note. This segment is the meta-point — the thing that separates this video from every other growth-role pitch the team has watched. Deliver it slowly. The line "falsifiable, not impressive" should land like a punchline.

Segment 6 — Closing Ask (3:40–4:00, ~50 words)

Visual: Webcam back to full frame. Black slide behind you with the closing line and your contact:

Let's spend 45 minutes co-editing the model
with your real numbers.

kaushtubhagrawal45@gmail.com

Talk track (look directly into the camera, no smile, calm):

"I've sent the four documents this video sits alongside. The strategy memo, the math memo, the operating plan, the meeting deck, and the day-one playbook. Read whichever altitude you prefer.

What I'm asking for is forty-five minutes of your time to co-edit the model live with your real numbers. That's the conversation I want to have. Looking forward to it."

Direction note. End with a beat of silence — half a second of you holding eye contact before the recording cuts. Resist the urge to wave or smile. The seriousness of the close is what makes the whole video read as a serious application.

Total runtime check

If you come in 30 seconds long, that's fine — the team will watch a tight 4:30. If you come in 30 seconds short, also fine — under-runs read as confident pacing. Anything over 5 minutes is too long for this format; cut Segment 3 if you need to.

Pre-recording checklist (90-minute session)

What NOT to do in the video

• Don't mention competitors by name. Same rule as the docs. If a comparison slips in, edit it out.
• Don't smile-and-wave at the camera in the cold open. Skip the niceties — your first sentence has to earn the next 4 minutes.
• Don't read from the script. Memorize the structure and the numbers. Speak the words. If you read, the team will hear it.
• Don't add background music. This is a peer-to-peer technical conversation, not a YouTube ad.
• Don't apologize for anything — pacing, the camera setup, the depth. Apologies in videos read as candidate-tone.
• Don't end with "thanks for watching." End with the ask. Close hard.
• Don't post this publicly. Privately to the team, alongside the docs. The strategy memo and math memo are the reading; this is the watching.

How to send it

When the video is recorded:

1. Upload to Loom (private link, no password — friction kills view rate)
2. Set the thumbnail to a frame from Segment 4 — the math memo on screen with your face in the corner. That's the most signal-rich preview frame.
3. Send the link with a one-line message: "Companion to the docs. 4 minutes. — Kaustubh"

That one-line message is doing real work. It signals respect for their time, sets the runtime expectation, and matches the operator-confident voice of the video. Don't write three paragraphs of preamble.

Why this video belongs in the package

The four documents are the depth. The deck is the meeting scaffolding. The video is the proof I can produce the kind of artifact I'm proposing as Play 3. A 4-minute customer-voice video with screen-share, real numbers, and a clean close is a case study — just one where the customer is me, talking about the role.

If the team watches this and thinks "this is what I'd want our customer case studies to feel like," the video has done its job. That recognition — that the artifact and the role's deliverable look the same — is the strongest possible signal that I understand what the role actually is.

— Kaustubh