The Workflow Company

We help B2B startups build and scale a consistent pipeline. We use AI and automation for precision in targeting, hyper-personalization in messaging, and ease in scaling.

Founder and CEO

First Round Fast Track

Mentor

Integromat

I help startups integrate Integromat into their operations to streamline and automate processes

Registered Partner

joerhew.com

I partner with early stage startup founders to bring clarity to their strategy and improve their operations. I help companies prioritize what’s truly important by leveraging data, then help implement a culture of quick iterations and learnings through no code. I like to get my hands dirty and execute, but also act as a strategic thought partner to the founders.

Clients include: Karrot (Daangn Market, Series D, backed by DST), Groove (Seed, backed by Resolute)

Fractional Head of Operations

Ops Hacks

Built a community of ops folks that work at early stage startups

Community Builder

Tyltgo (YC S20)

White-label, same-day delivery service for local merchants across Ontario

Co-founder & COO

Uber

(Sensitive figures redacted)

• Lead a team of 4 operations managers to perform continuous marketplace management of over 100+ municipalities across Canada with focus on growth, reliability, and efficiency
• Launched a series of experiments, features, and processes to consistently exceed targets on growth and costs - increased year-over-year deliveries by x% while reducing courier costs by y% 
• Provided managerial oversight to launch walkers as a delivery mode in Canada, which accounted for c. x% of deliveries in downtown Toronto within y months of launch

Head of Marketplace Operations, Uber Eats Canada

• Managed a cross-functional launch team for 3 months in leading the largest single-day launch in the history of Uber Eats across 30 cities in Canada, which increased national population coverage by c. x percentage points
• Built out Reporting & Performance Management function using SQL and GSheet, providing transparency and enabling accountability on organization’s OKRs for key stakeholders
• Conducted a technical analysis and modeling on x million+ historical transactions to determine the optimal pricing structure and project-managed the pilot of service fees in Canada

Business Operations, Uber Eats Canada

• Collaborated closely with the product team to launch a new carpool product, uberSHARE, the first P2P product since uberX the country's banning of uberX in March 2015
• Launched UberTRIP, an hourly rental product, to target tourists and business travelers and increased weekly city gross bookings by x% without incurring any additional costs
• Successfully overhauled the driver partner incentive program, saving costs equal to c. x% of weekly gross bookings on average with negligible impact on churn and engagement

Rides Country Lead, Uber Korea

Startups

1. One Hive: e-commerce app for international customers interested in purchasing Korean fashion items
• Supported UX and UI design of a mobile e-commerce service to market and sell Korean fashion items to international customers
• As business lead, gained 10K+ users and generated revenue within the first 6 months of launching

2. Foodle: mobile social networking platform with algorithmic recommendations for food enthusiasts 
• As product and business lead, designed UX of the mobile app and led marketing efforts across social and offline channels
• Collaborated with UI designer and front- and back-end programmers to launch product on iOS

Co-Founder / Business Lead

Morgan Stanley

• Built detailed financial valuation Excel models, based on assumptions and hypotheses derived from financial, operational, and economic data, to test feasibility and profitability of envisaged transactions
• Advised large Korean and multinational corporate clients on cross-border M&A transactions and other business development initiatives through providing financial and strategic analyses
• Selected transaction experiences include SK hynix’s US$ 248MM cross-border acquisition of Link_A_Media Devices and MBK Partners’ KRW 1.2Tn (c. US$ 1.0Bn) acquisition of a 30.9% equity stake in Woongjin Coway

Investment Banking Analyst

Citi

• Selected transaction experiences include serving as the Global Coordinator and Joint Bookrunner in Hyundai HCN’s KRW 83.6Bn (c. US$ 74MM) initial public offering and SK Networks’ US$ 700MM minority equity investment in MMX Mineração e Metálicos

We build automated workflows that discover, enrich, and execute on GTM intelligence across your entire stack - enabling your team to grow pipeline, not headcount.

Reach out to chat about AI <> GTM.

Tools in our stack we use on a daily basis:
- Supabase
- Node.js
- Render
- AirOps
- Attio
- Clay

We are a proud member of the 1% for the Planet and are committed to donating at minimum 1% of our annual revenues to help protect the planet.

Michelle on our team is not an engineer. 

But using Claude Code, she just automated herself out of a daily task - GTM campaign reporting for all of our clients.

What's more remarkable is that she literally completed this from start to end in less than 24 hours. 

I mentioned to her in passing that she could connect a couple sequencer MCPs to make her life easier on our daily sync yesterday. And this morning she sends me this message asking for access to a repo to push the code she and Claude Code co-wrote.

We adopted Claude Code + Github as the internal ops tool on Monday; and not 5 days later we have Michelle pushing code to automate internal reporting.

And again, I repeat - Michelle is very intelligent but does not come from a technical background. She is operations-oriented and she has been working with AI a lot over the past 1.5 years working at The Workflow Company, but setting up an automation like this would have been impossible without the help of AI.

It's absolutely wild.

This reporting task used to take Michelle 30-45 minutes a day - now she can spend this time doing something more strategic, creative, valuable and hopefully more fun. 

So yes, Michelle automated herself out of a task with AI. 
And no, she didn't lose her job over it - now she gets to work on more intellectually stimulating things.

My 2 cents - AI should be enabling increase in productivity and employee satisfaction; not headcount cost cuts and layoffs.

Joe Rhew

GTM Intelligence Orchestration | The Workflow Company

Just added back $576 to our bottom line + boosted our productivity by 3x at least at The Workflow Company.

What happened:

We've switched over our internal ops (wiki + project management + task tracking + internal comms) from Notion to Github after a 3 week pilot. 

Now all of our context - our meetings, our SOPs, our decision-making process, our content, our code - are contained in Github repos, issues, and projects. 

Add to the mix Claude Code, which can work alongside each of us with access to this context. 

Also add:
- Skills, that can follow your specific way of doing things with minimal upfront tokens to preserve the context window
- Agents, that have their own context window and can work on specialized and narrowly scoped tasks autonomously
- MCPs, that bring in standardized ways of interacting with other tools and more context

And it's looking like the best decision we've made for the company in a long time.

Bye Notion. Hi Github.

The attention economy is zero-sum. Here's why 80% of outbound fails before it starts.

Had a conversation yesterday that crystallized something I've been thinking about:

If you're doing what everyone else can easily do, you're competing for the same 20% of prospects that are already saturated with outreach.

Let me use Google Maps extraction as an example.

There are basically 4 levels of sophistication:

1️⃣ Level 1: Buy a database
You get the easiest-to-find businesses. They're in Apollo, ZoomInfo, every vendor list. They're exhausted. Desensitized. Your message dies in the noise.

2️⃣ Level 2: Use Clay's native Google Maps integration
Getting warmer. You can target specific cities, customize your search terms. But the thing is—in 2025, everyone serious about outbound is using Clay. You're still fishing in the same pond.

3️⃣ Level 3: Specialized data services (Serper, Bright Data)
Now you're getting warmer. More configuration options—zoom levels, zip codes, query permutations. You can get more specific. Most people stop here.

4️⃣ Level 4: Custom algorithmic extraction
This is where it gets interesting.

You write a custom algorithm that:
→ Tiles latitude/longitude coordinates across entire cities
→ Runs searches at each coordinate point
→ Doesn't stop until 95%+ of results at each point are duplicates
→ Intelligently balances thoroughness against diminishing returns
- Finds the long-tail businesses nobody else reaches

This isn't easy. You need to understand both the business strategy AND translate it into code that works.

But here's the tension:

Most companies testing a new channel think: "Let's try the quick way first. If it works, we'll invest more. And if it doesn't, we've limited our investment."

The problem with this approach is it becomes a self-fulfilling prophecy. If you only do what's easy, you get the same results everyone else gets. Then you conclude the channel doesn't work—when actually, you just didn't give it a fighting chance.

You're not just competing against your own efforts. You're competing against:
- Every other vendor reaching out
- Every automated sequence your prospects receive
- Every saturated list they're already on

The prospects you can easily find have already replied "f* off" 30 times this month.

The real question shouldn't be "Should we test quickly or invest deeply?"

Rather it's: 
"Are we being intellectually honest with ourselves with this experiment here? Are we properly testing whether this will work?"

At The Workflow Company, we focus on getting to the Level 4 systems. We'll start with a Level 2.5 to test quickly but understand we may not see good results until we get to Level 4. And therefore we are not for companies just looking to test quickly.

But if you want to build a sophisticated and bespoke system that fits your GTM needs - because that's where the alpha is - we might be the right fit.

Here's a GTM campaign idea for 2026 
(We've been running this since March and seeing good traction)

Contact-Based Landing Pages: a personalized landing page tailored to an individual person.

Yes, conceptually this might not be novel but the details make all the difference, and I'll share what we're doing.

1. Create a list(s)
🔑 Could be static, but trigger-based (website visit, new job posting) works really well. We use Clay for this

2. Enrich
🔑 Don't just enrich the target company and the person. Include in the prompt details about YOUR company and have LLM reason through how YOU solve THEIR problems. This part probably takes the longest because you want to iterate till you get it just right. We use Clay + Mastra (for the agent) for this

3. Create a landing page template in your CMS
🔑 Templatize the landing page - 80% static content + 20% AI-generated content tailored to the target. We use Webflow for this
🔑🔑 Create multiple templates for different personas on the buying committee

4. Create a personalized Opengraph image for the landing page
🔑 LinkedIn DMs will pull the OG image of links into the chat window to show a preview - use this real estate to catch the target's attention. We use Bannerbear for this
🔑🔑 We tried using AI image generation - the quality across 100s of pages wasn't consistent enough to push into production

5. Set up session replay and button clicks on the landing pages
🔑 Inject a pixel into the landing page template to monitor site visits, sessions (literally a video of the visitor scrolling and clicking around the page) and any specific button clicks you want to capture. We use Posthog for this
🔑🔑 You could install a site visitor pixel on the landing page for further validation. We use RB2B and Vector for this

6. Send on-page events in real-time to Slack
🔑 Create a Slack automation to alert the team any time a landing page is viewed or a CTA button is clicked. Follow up appropriately. We use Slack for this
🔑🔑 Attach to the Slack message buttons that trigger automated actions (e.g. enroll into a follow-up sequence)

7. Store the event data for future follow-ups
🔑 Store the data to re-use when you refresh your personalized landing page or simply to follow up in a few weeks. We use Attio and Supabase for this

8. Use the stored data for weekly analytics reports
🔑 Create SQL queries to pull data against your DB to track campaign performance. We use Supabase for this

This long process - especially steps 2 and 3 - really push you to be precise in your targeting. Great exercise to be doing regardless, and the quality of the output reflects the effort.

Have you done something like this already?

Most GTM leaders are asking the wrong question about AI.
"Does it save money, save time, or make money?"

The answer is yes. But not in the way you think.

Here's what I see companies getting wrong:

They pick ONE value prop for their AI strategy and try to force every use case into that bucket.

↳ "We're using AI to cut costs" (so they only automate existing roles)
↳ "We're using AI to save time" (so they only tackle admin work)
↳ "We're using AI to drive revenue" (so they chase the AI SDR dream and get burned)

This is like buying a car and only driving it to deliver food on Uber Eats.

The companies actually winning with AI in GTM are mapping different work streams to different value types.

Time-saving: Meeting notes, CRM hygiene, research that makes your reps want to quit

Cost-reduction: First-pass qualification, basic support, data analysis that used to need a full-time hire

Revenue-generation: This is where it gets interesting...

AI doesn't replace your best reps. It captures opportunities you're currently losing.

The follow-ups that slip through the cracks when deals get busy. 
And the long-tail accounts you don't have bandwidth to work. 
Or the re-engagement campaigns for stalled pipeline that never happen because your team is slammed.

These are pure incremental revenues sitting in your CRM right now.

But most companies never see it because they're too busy arguing about whether AI is a cost play or a revenue play.

It's both. And neither. Depends on the work stream.

So the real question isn't "What value does AI deliver?"

It's "For THIS specific workflow, what value should AI deliver, and how do we measure it?"

On context engineering for a copywriting agent.

I'm currently building a copywriting agent that takes a broad array of different contexts and distills them into a sequence of short messages.

There are 2 different ways of interacting with this agent. 

The first is a more rigid agentic workflow where we have a set of steps to collect and collate the contexts and feed them into the agent. 

The second is a more fluid, agentic chat interface where we don't know the user's intent ahead of time, and therefore cannot prep the contexts; instead the contexts have to be dynamically gathered by the agent.

For the first, we've kept the context composition very simple. We take all relevant bits of contexts like company overview, product overview, competitive positioning, etc. and feed all of them into the agent to output a sequence.

For the second, instead of pre-loading the agent with all the contexts, we've given it a tool to retrieve specific bits of context that the agent discovers it needs during the conversation.

Both of these more explicit context injections are also supported by a semantic similarity search layer, to augment with relevant memory/context retrieval.

A couple takeaways for myself and for anyone else interested:

- Copywriting is an interesting context engineering exercise, because the spectrum of context one needs feels very broad and deep, ranging from company overview to prospect details. Once you start building this yourself, you can see why the 2024 hype of AI SDRs was overblown and didn't deliver. I assume these companies have learned from trying to generalize and templatize everything too early and are approaching the problem in a more thoughtful, bespoke way

- The gap between building an agentic workflow (more rigid with some agency) and an agent is quite wide, and there are so many edge cases to think of and prepare for. Always best to start with a workflow and slowly expand

- That said, building tools and equipping agents with the ability to call them at their will is a trip. Adding a few lines of code at a time to "unlock" new abilities for your agents is a mind-expanding experience, and I get giddy at the sheer possibilities - even if we're not there yet

Is your lead scoring still stuck in the pre-AI era?

Traditional lead scoring gives you a number: "This lead is a 7 out of 10." or a "Medium Fit". Clean. Deterministic. Easy to route and prioritize.

But here's what I keep running into with clients: SDRs look at that "7" and have no idea what it actually means. The score works for sorting, but it fails at decision-making.

--
The observation:

Most scoring models combine database filters (headcount, industry) with some AI-generated attributes (intent signals, "strength of social media presence," engagement propensity). You get a weighted score.

But the rationale for the score is abstracted away. Your SDR sees a 4 and a 7, knows they should call the 7 first, but has zero context for how to approach either conversation.

What if lead scoring needs two layers instead of one?
↳ Quantitative score (the "7/10") - for routing and prioritization
↳ Qualitative context (the "why") - for understanding and action

Keep the first layer mostly  deterministic - company size, technographics, behavioral signals, AI-generated attributes, whatever your model weights.

The second layer is where AI actually helps. Not by making the score "better," but by explaining it with real data:

Example context block: 
Score: 7/10

Recent activity:
- CRO posted on LinkedIn yesterday about "evaluating new sales tools"
- Engineering lead attended our webinar 2 weeks ago

Company signals:
- Series B raised 6 months ago
- Hiring 3 SDR roles in past 30 days

Timing context:
- Q4 budget cycle likely starts in 2 weeks
- No demo requests but high research activity

Override signals:
- Engagement spike suggests urgency despite mid-tier score
- Multi-department interest (sales + eng) suggests internal testing

--
The shift this enables:

1. Agency - SDRs and agents can override when context reveals the score misses something
2. Transparency - Everyone sees the same reasoning
3. Better judgment calls - That 6-score lead who just posted about their pain point might be more valuable than the 7 who downloaded something 3 months ago

--
Future state thinking:

This context layer doesn't have to be static. Imagine the context is updated periodically and by real-time events.

And then you give an agent decision rights based on context thresholds:

"If a lead's engagement score spikes in a short period of time and they exhibit key buying signals, send personalized outreach."

The agent isn't making the scoring decision. It's acting on the combination of deterministic score + contextual signals that suggest the timing is right.

--
As we move to an era of abundant intelligence, we don't have to abstract away all the details and tokens. We have AI for that now. 

Ironically, we can now architect flows that feel less rigid and more human by removing humans from the process.

Anyone else experimenting with this? What am I missing?

Most GTM leaders are asking the wrong question about AI.

Yesterday I saw a CRO post: "What are the best AI tools for my AEs?"

This is a wrong question to be asking imho.

This is like asking "what's the best car?" without considering your personal context for needing a car.

Here's why this keeps happening:

GTM leaders approach AI in one of two ways:

1. "Tell me what AI tools to buy" (buying solutions to problems you haven't defined)
2. "Build me this AI system I spec'd out" (treating AI like deterministic software)

Both fail for the same reason: AI without strategy just automates mediocrity faster.

 ---
The right question: "What makes us impossible to compete with, and how do we 10x that?"

Here's the discovery framework:

Q1: "What makes you hard to compete against right now?"
- Can't answer? You don't have a strategy worth amplifying.
- "We're just better at execution"? That's not a moat.

Q2: "How are you doing that without AI currently?"
- Where are humans the constraint?
- Where are humans the magic?

Q3: "What happens if your competitors do the same thing?"
- Is your advantage structural or just timing?
- What's actually defensible?

Q4: "Where are the tradeoffs?"
- Cost vs. speed vs. quality
- Consistency vs. personalization
- Scale vs. depth

This is where it gets interesting. Because AI is non-deterministic, you can't predict these tradeoffs upfront:

Add more context → law of inverting returns
Give AI more agency → smarter but unpredictable
More guardrails → accurate but rigid

The right balance emerges through iteration.

Which means the "best tool" isn't the one with the most features. It's the one that gives you flexibility to iterate on the dimensions that matter to YOUR unique advantage.

If your competitive edge is deep personalization → you need tools that have deep integrations and extensive tools
If it's speed-to-response → you need tools optimized for latency, not depth
If it's consistency at scale → you need guardrails, not agency

The tool selection follows the strategy. Not the other way around.

---
Back to that CRO's question: "What are the best AI tools for my AEs?"

The better questions:

"What makes our sales motion impossible to copy?"
"How do our best AEs create that advantage today?"
"Where would AI amplify that vs. make us generic?"

Answer those first. Clarity on which tools to use will emerge naturally.

AI without strategy = expensive automation of the wrong things.
AI amplifying your unique strategy = compounding competitive advantage.

Which are you building?

I don't like the term "GTM Engineer". Never claimed I was one and never will. The role name has been co-opted by SaaS companies, and now it just means "I know how to use x, y, and z tools".

What I can get behind is "GTM Engineer-ING" - as a discipline, an approach and a way of thinking.

I'm excited to share my thoughts on this and an Attio work artifact today on the panel with my friends Andreas Wernicke and Alexandre Kantjas, hosted by Seth DeHart.

11AM Eastern on LinkedIn live. Register here:
https://lnkd.in/gzcKnCri

See you there?

Helping Reps book 20+ Meetings per month by targeting contact data that’s valid | Head of Sales | Prospeo.io

Go To Market Engineering: What Is It, and Do You Need It?

Hosted by Founder Led Sales and supported by Attio, this panel brings together Seth DeHart (Founder of FLS), Andreas Wernicke (GTM Consultant at Snowball), Alexandre Kantjas (Co-Founder at 9x), and Joe Rhew (Founder at The Workflow Company).

In this session, our experts will break down the concept of Go To Market (GTM) Engineering, a structured approach to designing, testing, and scaling the systems that drive customer acquisition and revenue growth. You will learn:

- What GTM Engineering actually is and how it differs from traditional go to market strategy

- When your business needs it and when it does not

- Key components of a strong GTM engine from process design to tooling and team enablement

- Real world examples of how startups and scale ups have leveraged GTM Engineering to accelerate growth

Whether you are a founder, sales leader, or growth strategist, you will walk away with practical insights you can apply immediately to optimise your go to market motion.

Olivier 🧪 Tytgat, Noëmie Jacquemin and 376 other attendees

GTM Engineering - What is it and do YOU need it? 

Taplio

Joe Rhew's Linkedin Analytics

What is Joe talking about?

Who is engaging with Joe

's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?