GTM engineering, explained badly

01A career announcement0:00

Narrator: For the last three months, I've been doing G T M engineering. Small career change. Same laptop. Different people disappointed in me.

Maya: Congratulations. What do you actually do?

Narrator: I connect... things. Commercially.

Maya: A plumber with LinkedIn.

Narrator: Please don't put that on my invoice. I used to explain why the software didn't work. Now the software works, and I explain why nobody bought anything. So the bugs have acquired purchasing authority.

Maya: And this conversation is going to explain your new job?

Narrator: Beautifully. You'll leave knowing when to nod, when to say interesting signal, and how to mute yourself before asking what pipeline means.

Maya: One real campaign observation. Everything else, including the travel plans, is a dramatization of a meeting that should have been stopped.

Narrator: The useful explanation is the next article. They can't make me do two jobs.

02The explanation escapes0:53

Narrator: G T M means go to market. You use data and software to help the right message reach the right person at a useful moment. There. That's actually most of it.

Maya: Oh! I got it. We're done.

Narrator: No, wait. I've booked the room for an hour.

Maya: Of course you have.

Narrator: The person belongs to an account, the account fits the I C P, the I C P has intent, intent becomes a signal, signal enters a workflow, workflow creates a task, task goes to David, David asks who this person is. Easy.

Maya: Where did David come from?

Narrator: Don't ask that. He came bundled with the C R M.

Maya: Is there documentation?

Narrator: There's a dashboard. It has a little loading circle. That's how you know it's thinking.

03A very specific everyone1:33

Maya: Let me try. Ideal customer profile. We decide which companies could actually benefit from the product.

Narrator: Yes. Our imaginary product makes meetings shorter. Who has meetings?

Maya: Everyone.

Narrator: Huge market. Write that down.

Maya: No. We need filters. Company size, relevant problem, whether they can buy it.

Narrator: More than fifty employees. Growing team. Head of operations. Public interest in efficiency, demonstrated during a forty-minute webinar about reducing unnecessary communication.

Maya: I attended that. There was a follow-up webinar.

Narrator: Now the spreadsheet is wider. Same companies, but you have to scroll sideways to doubt them.

Maya: I miss the plumber.

04Uncertainty gets dressed2:15

Narrator: Next: enrichment. The spreadsheet needs vitamins. We add websites, titles, company size. Useful information, allegedly.

Maya: This source says sixty employees. This one says two hundred.

Narrator: The website says small team, big mission. So: somewhere between sixty and a mission.

Maya: Which number goes into the system?

Narrator: We give it a confidence score. Uncertainty puts on a tie. Everyone relaxes.

Maya: Mine's wearing a waistcoat. Can it approve expenses?

Narrator: After enrichment, deduplication. David Smith, Dave Smith, D. Smith. We merge them. Three rows become one. Beautiful.

Maya: Three different people.

Narrator: Ah. We'll fix that downstream.

Maya: Where is downstream?

Narrator: Nobody knows. Excellent schools, apparently. All our problems move there.

05Fourteen points for a thumb3:05

Narrator: A signal is an event worth investigating. A new office. A relevant hire. Some context that makes a conversation timely.

Maya: Or someone liked a post.

Narrator: They liked a post about saving time. They want efficiency. They need our product. They're practically signing.

Maya: Their thumb slipped.

Narrator: Fourteen points.

Maya: For an accident? Why fourteen?

Narrator: Ten sounded made up.

Maya: What's the temperature of this extremely interested thumb?

Narrator: Fifty points: hot. Seventy: very hot. Ninety: David considers sending an email. The crowd goes... back to work, because nobody has actually spoken to a customer.

Maya: We've built a thermometer for a conversation that hasn't happened. Can it detect embarrassment?

Narrator: Enterprise plan.

06Human, according to three engineers3:49

Maya: Please tell me the message sounds normal.

Narrator: Hi David. Noticed you're Head of Operations at a company. As Head of Operations at a company, you probably care about operations.

Maya: You've caught him. He's operating.

Narrator: First draft. We improve the prompt. Add context. Remove hope this finds you well. New version: David. We know about the operations.

Maya: Come out with your workflows up.

Narrator: Put hope this finds you well back in. Immediately.

Maya: Three engineers and a language model inspect it. They agree a human could have written it. So we send it to a human.

Narrator: And?

Maya: Not interested.

Narrator: A reply! Engagement!

Maya: Stop celebrating!

Narrator: Don't tell the dashboard. It's having a good day.

07The actual observation4:35

Narrator: Here's the real observation. In a LinkedIn campaign, our women business development representatives got roughly twice the connection acceptance rate of our men. Business development representatives are the people doing the outreach.

Maya: That's a result worth looking into. Were the audiences comparable? Messages? Profiles? Timing?

Narrator: Exactly the questions you'd need. The result alone doesn't tell us why. It isn't a rule about women and men.

Maya: Good. So what did the fictional meeting inside your head recommend?

Narrator: Thailand.

Maya: Sorry. What?

Narrator: Next year's offsite. Thailand. Two bars on a chart. One airplane. A surprisingly short presentation.

Maya: You haven't controlled a single variable, and you're choosing seat numbers.

Narrator: David and Sean asked what this had to do with team building. I said: everything, gentlemen. We are becoming data-driven.

Maya: Somebody take the laser pointer away from him.

08An extremely fictional offsite5:30

Narrator: Entirely imaginary offsite. Costume party. Sunshine. A slide deck with no return ticket to common sense. David and Sean have negotiated the wardrobe budget before anyone has tested the idea.

Maya: Their glamorous lady alter egos look incredible. The wigs. The outfits. The exact same unresolved C R M permissions.

Narrator: David wants to know if the new persona gets a higher base salary. Sean has submitted false eyelashes under sales enablement.

Maya: Finance approved one eyelash. The other one's a stretch goal.

Narrator: No. Don't put that in the deck. They'll actually do it.

Maya: The women on the team have a cheaper suggestion. Test the targeting. Test the message. Same people, better experiment.

Narrator: Too late. Leadership has seen the airplane. You cannot unshow an airplane.

Maya: Everyone keeps their identity. Nobody changes anything except outfits. The joke is your business reasoning, which has somehow become the least convincing costume.

Narrator: The wigs have better methodology than the deck. I admit that. But mine has transitions.

09A small problem with revenue6:37

Maya: Did the extra connections become useful conversations?

Narrator: We were celebrating.

Maya: Accepted connection. Meeting. Customer. Happy customer. Different things. Someone still has to want the product.

Narrator: A dependency on the product. At this stage?

Maya: Where are those numbers on the dashboard?

Narrator: Under Advanced. Collapsed by default. The first number is very large and green. That one's doing wonderful things for morale.

Maya: Revenue is an advanced feature.

Narrator: We've already paid for the costumes. We're emotionally invested in the hypothesis. Finance doesn't have a column for that, which is why it's so expensive.

Maya: Downstream has a column.

Narrator: Downstream has everything.

10The doorbell architecture7:18

Narrator: Finally. Automation. Something I recognize. A webhook is one system telling another system something happened. Like a doorbell.

Maya: Ding dong.

Narrator: That rings another doorbell. That updates a spreadsheet. That sends a message saying the original doorbell is unavailable. Someone retries it. Now there are four doorbells.

Maya: Ding dong!

Narrator: Please. I used to sleep. We've got rate limits, missing fields, and a timestamp arriving from tomorrow to warn us.

Maya: A human checks the output before anything embarrassing goes out, yes?

Narrator: Yes. Even this fictional company can afford that. Signal, enrichment, score, owner, task, human review. There is a useful system in here. It's just surrounded.

Maya: Time saved?

Narrator: Six minutes. Took an afternoon to maintain. But next week, thousands of runs. Potentially worth it. Potentially my new weekend.

Maya: The diagram doesn't fit on my screen.

Narrator: Architecture.

11One invoice, four victories8:15

Maya: Someone bought the meeting software!

Narrator: Sales says it was the call. Marketing says the campaign. Product says the product. I say the integration, because I've been here until eleven.

Maya: The customer says a colleague recommended it.

Narrator: Does the colleague have an A P I?

Maya: First touch gets credit for the first interaction. Last touch gets the last. Which are we using?

Narrator: Our fictional all-touch model. Everybody wins. Four customers from one invoice. Finally, a conversion rate that respects the team.

Maya: One customer.

Narrator: One customer. Four departments. I connect the systems so we can see what happened. Then they disagree and I become a diplomat with A P I keys.

Maya: You wanted a change from debugging.

Narrator: I debug the sentence this lead should have been mine. The sentence has been in production since 1998.

12Certification pending9:03

Maya: Hang on. Find suitable companies. Understand them. Notice useful moments. Build reliable plumbing. Check whether it helped. That's the job?

Narrator: You're getting dangerously close. Let me help.

Narrator: We operationalize an intent-led, signal-enriched, multi-touch revenue motion across the entire customer lifecycle, with a human in the loop and Sean in a very good wig.

Maya: I had it. I actually had it for a second, and now it's gone.

Narrator: Perfect. You're ready for the kickoff.

Maya: What do I do if someone asks about revenue?

Narrator: Dashboard's refreshing.

Maya: Thailand?

Narrator: Downstream.

Maya: David?

Narrator: Merged with two other Davids. We're sending flowers.

Maya: Write the useful version, please. I would read it. I might even understand what you do.

Narrator: I will. For now, three months in, one solid conclusion: test the message before booking the flight.

Maya: Interesting signal.

13Until next time9:57

Narrator: Thanks for listening. Follow along for more articles, experiments, and things that worked on my machine.

Maya: Some of them might even work on yours.

Narrator: No promises. Have a lovely day. Go make something.

Maya: Preferably a cup of tea first. Bye!