Automate, Delegate, or Kill? The Truth Behind Task Automations

Summary

Podcast Summary: Special Ops with Emma Rainville

Episode: Automate, Delegate, or Kill? The Truth Behind Task Automations
Date: October 28, 2025
Host: Emma Rainville (Guest hosting by Richard & Tiago from Shockwave)

---

Episode Overview

This episode of Special Ops takes a tactical deep-dive into the critical business question: Which tasks should you automate, delegate, or eliminate ("kill")? Regular host Emma Rainville hands the mic to Richard and Tiago from the Shockwave team, who unpack practical strategies—using real business examples—to help founders and operations leaders make their teams more efficient. The aim is to clarify how to decide among automating, delegating, or killing a task, so that your most valuable resource—time—is spent with maximum impact.

---

Key Discussion Points & Insights

1. The Double-Edged Sword of Automation

• Automation Isn’t Always the Answer:

  • Richard: "Automation on some level takes information away from you. That can be a very good thing, that can also be a very bad thing." (00:00)
  • Automating tasks means you lose a tactile sense of what’s really happening. For critical areas (like daily workshop signups and ad performance), the hands-on approach can yield actionable insights that automation might obscure.

• Criteria for Automating a Task:

  • Tiago: Three critical questions to ask before automating:
    1. Does the task truly need to be done?
    2. Does it require standardized execution?
    3. Is the task frequent enough to justify the setup time?
       (04:00)

• Good Automation Subjects:

  • Low-variation, high-frequency tasks like email deliverability checks.
  • Tedious processes assigned to junior staff, e.g., manual copy-pasting between apps or systems.

---

2. Delegation: Moving Tasks to the Right People

• Delegation vs. Automation:

  • When automation can’t handle the cognitive, context-specific, or customer-centric aspects of a job, delegation is the next move.
  • Richard: "If you don't have those, that kind of hands-on view, stuff gets away from you. It's just the nature of the thing." (03:50)

• How to Decide What to Delegate:

  • Task does not require the founder’s unique skills or experience.
  • Task can be systematized, standardized, and monitored by others.
  • Prioritize delegating tasks that are simple but consume a lot of time due to their frequency.

• Practical Exercise:

  • Make an “accountability chart.”
    • List every responsibility currently handled by the founder or manager.
    • Identify which roles could be hired or developed to take tasks off your plate.
    • Start delegating high-frequency, low-complexity tasks to reduce both workload and context switching.
      (10:02)

---

3. Killing: Challenging Traditions and Redundant Work

• Eliminating Unnecessary Tasks:

  • Richard: "How easy it is to build up these traditions that don't really need to be followed at all. Like, okay, why am I doing this? Well, I've always done it." (11:35)
  • Regular business review or bringing in an outside perspective (consultants, new hires) gives the distance needed to identify legacy processes that make no current sense.

• The Value of Perspective:

  • Tiago: “If you can get more time for you, you'll also get the ability to change the perspective because you have time to think at a different pace and quality and analyze.” (12:17)

---

Notable Quotes & Memorable Moments

• On the Purpose of Automation/Delegation:

  • Richard: “The point of automation, the point of delegation, the point of all of this is to free up valuable time and push that to places where it can be more efficiently used.” (07:02)

• Real-World Example:

  • Workshop registrations: Previously handled in detail by the owner; after mapping and selectively automating processes, plus delegation, the owner now has “a lot more thinking free time.” (08:41)

• Founders Limiting Themselves:

  • Tiago: “It won't make sense once you write it down...” (10:02)
    (Referring to the moment a founder lists out all their roles and realizes how fragmented their time is.)

• On Building Accountable Teams:

  • Start by delegating entire contexts or related clusters of tasks to minimize context-switch costs, rather than just one task at a time.

---

Timestamps for Key Segments

• 00:00-03:17 – The risks and reasoning behind automation; when too much distance is a problem
• 04:00-05:46 – Tiago’s three questions for automation; evaluating task complexity and frequency
• 07:02-08:07 – Delegation: what to move off your plate and how to structure escalation
• 08:41-10:02 – Founder’s self-assessment: accountability charts and context-switch elimination
• 11:35-12:17 – Killing: the power of periodic process review and letting “traditions” die
• 12:17-12:49 – The importance of outside perspective and time to think

---

Final Takeaways

• Use automation for repetitive, standardized, high-frequency tasks—but beware the loss of firsthand insight.
• Delegate anything that doesn’t require your unique skill—and focus on systematizing and monitoring, not just offloading.
• Routinely review ongoing tasks and kill those that are only alive by inertia; an outside perspective helps.
• Always tie back to the goal: freeing up your best people (including you) for the highest-impact work.

---

This summary captures key ideas, processes, and actionable frameworks as discussed by Richard and Tiago, preserving their practical, operation-focused approach—and providing founders and managers tactical clarity on automating, delegating, and eliminating tasks.

Transcript

[Richard]

Automation on some level takes information away from you. That can be a very good thing, that can also be a very bad thing. And that's really what I try to break down there.

[Tiago]

You have to pick what's good to automate.

[Richard]

The point of automation, the point of delegation, the point of all of this is to free up valuable time killing stuff. It really does require an external perspective. If there's no time to answer that, there's probably something to be automated.

[Tiago]

I was thinking about probably three things that you need to ask yourself to make sure X task should be automated. So, first of all.

[Podcast Host]

Hi and welcome to another episode of Special Ops podcast. Today we have a special episode for you because Richard and Tiago from Shockwave's team are actually going to do the podcast for you today. They're going to be talking about what to automate, what to delegate and what to kill. I'm going to sit this one out because all I do is throw chaos at them anyway. So they're going to tell you how to simplify all of that. So join them. Here you go.

[Richard]

All right.

[Tiago]

So, Richard, so happy to be here with you. An amazing experience to meet you and to meet Emma here in San Diego. So I was remembering the first time I met you. I think it was not the first time, but when I was being interviewed by Emma to join Shockwave, it was almost a year and a half ago and I was talking through what I knew, what I had been doing for another company and she just post and she said, I need to have Richard here and I need to bring Richard to ask you some questions. Do you remember that time?

[Richard]

I don't remember as well as you died. I apologize, I've ruined your intro there.

[Tiago]

I really recalled. No, no worries. I recalled that you came onto the interview to ask me about what I knew about Automations. We talked about again a year and a half away and now it's been more than a year. Working together, working closely together, which I love and I learn a lot from you. And, and one of the things I really value about you is your criteria on what to automate, what not to automate, what to keep manual, mostly. So I wanted to ask you about again, how does that work for you on your head? What's your criteria?

[Richard]

Well, I mean, I think the fundamental thing is automation on some level takes information away from you, which that can be a very good thing. That can also be a very bad thing. And that's really what I try to break down there. So there are a lot of daily reports I could Completely automate, for instance, like the number of sales for a workshop that we've got coming up. We have this big sheet for one of our clients, the one that we work most closely together on, which is basically how many people have signed up for this upcoming workshop. I go in there manually every day, I update the spend, I update the number of people who've come in. But if I automated that, that would take away the ability to say, okay, are we doing well for this consultation? Are we having a reasonable number of signups? Are these new ads that we've launched actually working or is there some kind of issue that we need to look into? The automation there would just completely kill. That would add a step between the information and our understanding of it. I think that's the sort of core criteria for me there. Whereas something that's a lot more automatable is for instance email deliverability rates. That's something that doesn't change super quickly unless you're doing something really bad. That's something that can be automated and occasionally checked into because the proximity of the information is less important than it is for the number of consultation signups. That's definitely my kind of criteria there.

[Tiago]

There are risks associated with automating. I hadn't thought about that specifically or explicitly, which is, hey, you're going to forget about it. So that's part of the good part and it could be, as you said, a negative part at risk. You're not going to look into that as often as. Sure. And I think it happens with both automations and it happened with AI. And I know I have you on my side where someone goes like, oh, I can automate this, I can automate that, I can automate the other thing. You can certainly do same with AI, like oh, AI could do this. You have to pick what's good to automate. Absolutely, definitely.

[Richard]

Because if you don't have those, that kind of hands on view stuff gets away from you. It's just the nature of the thing. I think there's some level of truth in that with regards to delegation, but we'll get into that in a second.

[Tiago]

So I was thinking about automating and I was thinking about probably three things that you need to ask yourself to make sure an X task should be automated. First of all, it's very simple. It needs to be done. So it's not something that we could do. It needs to be done because automated is going to just do. It needs to be done right in a specific standardized way. It has to be done in a timely manner because automation will always be human being timely and then, and I think this goes back to the criterion on whether they should also allocate time to set up the automation. It has to be a task that has to be done frequently because if it's just you have two employees and you have payroll once a month, twice a month, there's probably something better that you could automate that would relieve you of a lot of workload.

[Richard]

Yeah, I mean understanding whether time is being spent is I think something that you've done fantastically well. Again for this client we won't name but like just understanding where their staff has been spending that time because a tremendous, tremendous amount of it was automatable in a productive way that it drags everything forward rather than adding any roadblocks there.

[Tiago]

Yeah. And I think that automation is really connected to delegation, which you just hinted. We're going this way because when you think about the organizational chart and the hierarchy there, when you think about, and I'm calling it lower level team members, but not because it's low what they do, they're vital. It's about like just the vertical hierarchy. Right. So lower level are the ones that can get the most relief out of automations immediately because it's probably who's been assigned the task to manually copy and paste things from one side to the other. You know, hey, or even worse than copy and pasting, I think it's, hey, you should remember three times a day to go into the CRM and check if we have new subscription and then copy and paste information to this checklist.

[Richard]

Yeah.

[Tiago]

Now the moment we automate that for them and that goes immediately into the checklist, it's my most reliable and everything. And they get an automatic notification via Slack that they see and that everyone can see. So it's now like public. We are giving them more time to focus on tasks that really need a human input that cannot be delegated to machines. So I think first of all delegate to the machine. And after that when we say okay, we delegated a lot to the machines and now we think about a again, production team members are being relieved of tasks. You as a manager, in your experience, what's your criteria again to decide? Okay, what can we now delegate from managers into production?

[Podcast Host]

Yo, we interrupt this POD to tell you like and subscribe. What are you doing? Why haven't you liked? Why haven't you subscribed? Just subscribe. What's the problem? In all seriousness, subscribe so that you get notifications every time we drop new content. Additionally if you have not signed up for our visionary vault, what the hell. Www.specialopspodcast.com Go sign up. It's free. We never try and sell you and we're putting all kinds of stuff in there to help you with the operations of your business because we're passionate about it and we want to share operational excellence with our direct response, e commerce and online selling family.

[Richard]

Sure. So I think one of the core things there is the level of connection to the customer. Like in general terms, if you are again on that hierarchical level higher up, it's probably not going to be a great use of your time talking to customers, even if they are relatively high ticket. So just as an example. So that's something that can be delegated with a structure for actually escalating the stuff that does need to come to someone higher up the totem pole, really. It's a case of understanding what can be safely freed up. Again, as he said, the point of automation, the point of delegation, the point of all of this is to free up valuable time and push that to places where it can be more efficiently used. Like people who are a bit lower down, who are able to run through those little tasks, or not necessarily less important tasks, but less cognitive tasks, I think is possibly a good way to put it. Like stuff that takes less active thought or it needs to be done in a way that's not nowhere near automatic, but not necessarily needing a huge amount of thought, if that makes sense. Yeah.

[Tiago]

So I'm trying to put my, my, myself back in our day to day. Right. And again, usually a founder who's handling a lot of tasks and we're trying to get him or her to delegate, you're looking at what to delegate. So what are the things that don't demand your unique special experience and your forte, let's say, what are the things that you could train someone on? And you allocate a bit of your time into building the training for that and then you add a bit of monitoring cadence to make sure that's been delegated properly.

[Richard]

Sure. I think again, not to keep using the same client as an example for everything, but I think it's a very good example. The owner originally handling basically everything around checking people in for this product that they have. They were handling the actual mechanical side of signing people in for that, getting them registered, sending out notifications like, hey, there's stuff that the information that you've given us doesn't quite match our requirements. So we're going to need you to redo this. It was Taking, as you know extremely well, a tremendous amount of their time. You have automated a huge part of that. You've delegated the pieces that can't be immediately automated. And the person, the owner, now has a lot more thinking free time. I think that really is a criteria. Like, is this something that really has to be done? What's gained by you, the owner, or you, the higher up person, doing it? If it's nothing, then it's almost certainly delegatable. If it's, hey, there's a slight edge, it's a little better when I do it versus when someone else does it. Is that a trainable edge or is that just a question of figuring out how to approach the task? Because I think from your conversations, a lot of that has been now completely passed off. Yes, in a way that today, say it's better. I'm not intimately connected enough to say, is it running better? It's certainly running as well, and it's certainly freeing up that time.

[Tiago]

You have to look as a manager or a founder before delegating all the tasks you're doing right now. If we want to think about kill, then that's a good point to ask yourself, like, before delegating, should we in general be allocating time in general from the company? Could that be killed? But to delegate, make the sort of accountability chart that a visionary founder who's not yet very organized needs, needs to do, which is like, it won't make sense once you write it down is like, okay, I'm doing fulfillment, I'm doing cs, I'm doing finance, I'm doing payroll, I'm doing marketing as well. So list everything you have, look at your team and look at the empty spaces, who you could hire. And then start from the simplest task and probably do a little math where it's like, okay, this task is simple and how many times a week, a day I need to work on this as a founder or a manager, Right. So multiply that and you can start prioritizing which you should be delegating first. And I think you could also combine that with context. Hey, I'm doing things in five departments. Maybe I can focus on just delegating for two departments, building up customer support and building up, let's say marketing, for example. And then I'm also going to relieve myself from that context switching. So two different ways. Just try to relieve yourself from specific context and delegate that. Or try to go about, okay, tasks that are fast, simple, but are done multiple times every day, week. And then you can select those to delegate so any final note regarding automation, delegation and killing tasks?

[Richard]

I think killing, just. Just to have a quick note there, interesting thing there is how easy it is to build up these traditions that don't really need to be followed at all. Like, okay, why am I doing this? Well, I've always done it. Why are we doing this as a company? While we've always done it, it made sense two years ago, it made sense three years ago. I think killing stuff, it really does require in some cases an external perspective or that thinking time. I mean, a lot of the time, again, going back to the exact same client, there are a lot of things now that the owner has this additional time, they've been able to say, okay, well, why were we doing this from the beginning? This doesn't make sense for our. I mean, because if there's no time to answer that, there's probably something to be automated, probably something allocated, and then you can figure out where the rest of the time goes.

[Tiago]

And as you said, it has to do with perspective as well. Either if it's like an outside party, if it's a new employee, probably hire comes fresh, or a consultant, of course you're going to pick Shockwave. And if you got that perspective change from outer perspective, or if you can get more time for you, you'll also get the ability to change the perspective because you have time to think at a different pace and quality and analyze. And then the question, as you said, that's going to come up is going to be, why do this? Okay, Richard, thank you very much. It was a huge pleasure.

[Richard]

It's been a fantastic little show.

[Podcast Host]

I really hope you liked today's episode as much as I liked watching it from the sidelines. These guys are brilliant and I hope you understand why I talk so highly of them on the podcast all the time. We've created for you Shockwaves custom GPT that will help you decide. Do you automate? Do you delegate or do you kill?