Claude vs ChatGPT vs Lovable – For Partnerships

Published on April 2026
Expert advice from Rob Moyer (Founder, Bluethread.io) and Justin Zimmerman (Founder, Partnerplaybooks)

Table of Contents

Snapshot

You are not too late to AI in partnerships, but you are late if you are still waiting for a perfect tool, a perfect policy, or a perfect internal rollout. The big shift happening right now is all about which teams can move faster, share knowledge better, and turn scattered information into action. That matters a lot in partnerships, where speed, clarity, and responsiveness often decide whether a deal moves forward or dies quietly.

Justin and Rob make a practical case for a different approach. Start with the tools your company already allows. Learn one deeply. Build around your own data. Keep a human in the loop. Then use AI not as a gimmick, but as a way to solve real workflow problems like enablement, co-sell support, messaging, partner communications, and internal knowledge sharing.

If you want to solve tool confusion, workflow bottlenecks, and slow partner enablement, keep reading to see how Justin Zimmerman and Rob Moyer can help you do it.

“Always be learning. Scratch that curiosity itch.” -Rob Moyer

Why this conversation matters

If you work in partnerships, alliances, ecosystem, channel, or co-sell, you are probably dealing with the same mess everyone else is dealing with.

There are too many AI tools. The rankings change every month. One person swears by ChatGPT. Another says Claude is better for thinking. Someone else is deep into Gemini because their company runs on Google Workspace. Then a new tool shows up and suddenly everyone is posting screenshots like they discovered fire.

That chaos creates two bad outcomes.

  • You freeze and do nothing because the options feel overwhelming.
  • You bounce from tool to tool without ever building a repeatable workflow.

Justin framed that tension perfectly. For a lot of people, the problem is not lack of interest. It is not knowing where to start, what each model is actually good for, and whether switching tools means losing all the context you invested in the last one.

Rob’s answer is refreshingly grounded. Don’t start by chasing hype. Start by mastering the tool in front of you and meeting your customers where they are.

That means asking simple questions first:

  • Are you in a Claude shop?
  • Are you in a ChatGPT shop?
  • Are you in a Gemini and Google Workspace environment?
  • Are you in a Microsoft Teams and Copilot environment?

Your best AI tool is not always the internet’s favorite AI tool. It is often the one your company approves, your team can access, and your workflows can actually use.

This is especially true in partnerships, where your day is rarely about one giant task. It is about a hundred small actions:

  • answering partner questions quickly
  • building enablement materials
  • creating messaging for sales and partners
  • organizing knowledge
  • translating strategy into something your team can use

That is why this conversation matters now. AI is quickly becoming less about novelty and more about operational advantage.

“What’s everybody’s number one AI tool that they’re either fanatical about or deep dive using right now?” -Justin Zimmerman

How to choose an AI tool without overthinking it

Rob’s first principle is one of the most useful in this entire discussion: master the tool you are already using.

That does not mean you should never experiment. It means you should stop acting like the perfect tool will save you from shallow usage.

These platforms are leapfrogging each other roughly every quarter. Claude may be having a moment, then ChatGPT improves. Gemini gets better. Copilot closes a gap. New products show up with cleaner interfaces or more specific workflow support. If you build your whole strategy around hype cycles, you will spend more time switching than learning.

A more durable way to choose looks like this:

  1. Start with your approved stack. If your company is already standardized on Google or Microsoft, that matters.
  2. Identify your workflow problem. Are you brainstorming, building decks, training teams, answering partner questions, or creating internal knowledge systems?
  3. Use the best tool for that specific job. Do not expect one tool to dominate every use case.
  4. Keep your source material portable. Your real advantage is your data and how you organize it.
  5. Use human judgment on final output. The best results still come from discernment, not blind acceptance.

This lines up with what many partner teams are already discovering. AI becomes useful when it plugs into your daily motion, not when it sits off to the side as a toy.

If you want a broader strategic lens on where this is heading for ecosystem and revenue teams, this practical playbook on how AI will reshape partnerships is a strong companion read.

What each tool is good at right now

Rob did not declare a single winner, and that is exactly the right answer.

Still, he drew some useful distinctions that can help you decide where to spend your time.

ChatGPT

Rob described ChatGPT as the most creative brainstorming app. That tracks for a lot of people. It is often strong when you need:

  • idea generation
  • concept development
  • creative reframing
  • initial drafting
  • fast exploratory thinking

If you are whiteboarding messaging, drafting a partner pitch, or pressure-testing campaign concepts, ChatGPT is often a strong starting point.

Claude

Claude is the tool Rob has been leaning into recently. He likes it, but he also pointed out that its behavior has changed over time. In his experience, Claude now pushes back a little more than it did before. That can be useful if you want stronger reasoning or a model that does not simply agree with whatever you ask.

Claude has become a favorite for many people doing:

  • strategy drafting
  • document work
  • structured thinking
  • long-form writing support
  • workflow building alongside tools like Lovable

Gemini and NotebookLM

Rob made a point that many teams miss: do not sleep on Google.

For partner teams already living in Google Workspace, Gemini and NotebookLM can be especially practical. NotebookLM stood out in the conversation because it helps reduce hallucinations by grounding output in the materials you upload.

That is a big deal for partnerships, where accuracy matters and where your team is often trying to answer specific questions based on partner program docs, product information, and internal enablement material.

Rob shared that he often creates a dedicated NotebookLM environment around partner data and then uses that as a shared brain. Because it can be shared, it becomes useful not just for one person but for a team.

Copilot

Copilot came up less as a personal favorite and more as a practical choice inside enterprise environments. If your organization already runs heavily on Microsoft Teams and related tools, Copilot may simply be the path of least resistance and highest compliance.

Lovable

Lovable was the wildcard in the conversation, and maybe the most interesting one.

Rob described Lovable as something that “hits the easy button” for building decks, presentations, and thought leadership assets quickly. But he is not just using it for one-off content generation. He has moved into using it as a design engine for publishing client-facing workflows and tools.

“I actually have ChatGPT in one and Claude in the other, and I’ll type the same prompt and watch what comes from it.” -Justin Zimmerman

Stop obsessing over prompts

One of the smartest ideas Rob shared is that it is no longer mainly about writing the perfect prompt.

That may sound odd in a world still obsessed with prompt hacks, but his point is simple. The better move is to explain the outcome you want and let the tool help you shape the request.

Instead of this:

  • Write me a prompt that creates a partner enablement deck

Try this:

  • I need to create a partner enablement deck for a co-sell audience. I want it to be practical, clear, and specific to this product. Help me write the best prompt to get there.

That shift matters because it moves you from “chatbot user” to “workflow operator.”

You are no longer trying to be a magician with syntax. You are telling the system what success looks like.

Rob also emphasized another important practice: get multiple opinions. He will compare outputs from Gemini and Claude, then use his own judgment to decide what is best. That human-in-the-loop discernment is, in his words, the difference between okay output and great output.

That is an important reminder for any team nervous about AI replacing expertise. In most real-world partner work, the value still comes from the person who knows the audience, understands the relationship, and can judge whether an answer is actually useful.

How to switch tools without losing your brain

This is one of the hardest practical problems in AI adoption.

You invest weeks or months into one model. It learns your preferences. It starts sounding more like you. It remembers patterns in your work. Then another tool becomes more attractive for a certain use case, and suddenly switching feels expensive.

Justin called this out directly. The more you invest into a tool, the harder it becomes to leave because the responses feel more tailored and the relationship feels more developed.

Rob’s answer was elegant: do not let any one model become your actual brain.

Instead, build your own external source of truth.

He has done this by organizing his documents, ideas, and working material inside Google Drive. He even gave it a name: the Blue Thread Brain. Because his thinking lives in a central repository rather than inside one model’s memory, he can point different tools at that repository and get much of the same value.

His logic is worth adopting:

  • If your documents and thinking are stored in one place, any model can access them.
  • If any model can access them, switching tools becomes easier.
  • If switching tools becomes easier, you can choose based on workflow fit instead of emotional attachment.

Rob estimated that pointing a tool to the right database gets him about 80 percent of the way there. The remaining 20 percent is nuance, and that can be trained through instructions like tone, style, or formatting preferences.

This is a powerful pattern for partner teams. Your knowledge base should outlast any one AI vendor.

If you are thinking about AI as part of a larger revenue and data strategy, this piece on partnerships, data, and AI for revenue teams reinforces the same idea: clean, portable data matters more than tool hype.

“If I have my repository of documents and thoughts in one place, I can point whatever tool I’m using and it’ll get 80 percent of the way I’m thinking.” -Rob Moyer

Why Lovable clicked for Rob

Rob’s move from Notion toward Lovable came down to usability and client experience.

He told a very relatable story. He had an event in New York and was stressed about building the deck. He knew Claude could help build content, but he wanted something more customized. A friend mentioned Lovable. What he expected to be a two-day project ended up taking two hours.

That was the unlock.

Then he started asking the obvious follow-up question: what else can this do?

His answer was bigger than slides.

He moved his website into Lovable and began using it to create client-facing workflow tools. He described himself as “pseudo-technical,” someone who knows enough to build, push to GitHub, and deploy with Vercel. Lovable became the design engine that made those workflows easier to create and easier for clients to use.

That solved a problem Notion was not solving as cleanly. Notion gave him a strong database, but sharing with clients involved more friction. Lovable let him turn the same knowledge into a public, usable interface.

That distinction matters. A lot of internal AI experimentation dies because the output is technically correct but practically clunky. If the end experience is confusing, people will not use it.

Rob’s takeaway was not “everyone should leave Notion.” It was more specific: choose the interface that reduces friction for the person receiving the information.

That is a partnership principle as much as a product principle.

What partner teams should actually build

This is where the conversation got especially useful for operators.

Rob shared examples of what he is actually building in Lovable for clients and for his own work. These are not vague AI experiments. They are practical workflow assets.

He described a set of client-facing tools that function like workflow trainings. Examples included:

  • co-sell operating frameworks
  • tech stack guidance
  • deal calculators
  • brief templates
  • revenue AI playbooks
  • GTM health checks
  • enablement resources

The point is not that every partner manager should suddenly become a product builder. The point is that AI now makes it possible for small teams to package their knowledge into something more usable than a spreadsheet or static document.

That is a major shift.

For years, a lot of partnership knowledge has lived in places like:

  • random docs
  • one-off slide decks
  • buried wiki pages
  • tribal knowledge inside someone’s head

AI-assisted tools can turn that into living enablement.

Rob also made an important distinction about his intent. He is not trying to trap clients inside his own site. He wants them to learn what is possible so they can build versions for themselves.

That is the right mindset. The best partner operators are not just moving work faster. They are creating reusable systems.

“Think of it this way. I built some of these that are basically workflow trainings.” -Rob Moyer

From system of record to system of action

One of the strongest frameworks Rob shared was this split between systems of record and systems of action.

Here is the basic idea:

  • Your CRM is your system of record.
  • Your AI layer often becomes your system of action.

That means your CRM, PRM, and core business systems still matter. They hold the official data. They track the history. They define what happened.

But AI can help you do something with that information faster.

It can help you:

  • extract insights
  • create messaging
  • support reps and partners
  • produce enablement
  • respond to questions in real time

This is a healthier way to think about AI than the usual build-versus-buy debate.

Rob said he does not think about build or buy only at the company level. He also thinks about it at the individual level. Use AI to optimize your life. Use the tools your company provides. Master those. Then add lightweight systems of action around them.

That creates a useful middle path between two extremes:

  • waiting for a giant enterprise AI rollout
  • going fully rogue with disconnected tools

You can stay grounded in your system of record while still creating faster, more useful actions around the edges.

The rise of small, fast team AI

There was a subtle but important tension in this discussion.

On one side, large organizations need approval, governance, procurement, and cross-functional alignment for major AI deployments. That is real. If you are implementing AI across multiple departments or wiring it into deep workflows and sensitive data, you need process.

On the other side, small teams often understand their own pain points better than anyone else and can move much faster with lightweight solutions.

Rob is seeing more partner teams work together to build focused internal tools that fill the gap between corporate platforms and day-to-day execution. These are not massive platforms. They are narrowly useful tools for things like:

  • fast partner enablement
  • product messaging
  • sales support
  • internal training
  • shared knowledge access

Often these are built in approved environments like ChatGPT, Claude, NotebookLM, Gemini, or Copilot. The team trains the tool on its product, partner motion, or messaging framework, then uses it to respond faster.

This is where AI gets very practical for partnerships.

A partner manager’s job is often constrained not by lack of expertise, but by the time it takes to turn expertise into something another person can use. AI shortens that distance.

That means the advantage is not just intelligence. It is responsiveness.

And in co-sell, alliances, and ecosystem work, responsiveness is often what builds trust.

“As you’re a partner manager, one of the things AI does is it gives you the ability to react faster and get information to people faster.” -Rob Moyer

What this means for co-sell and ecosystem leaders

Rob’s final examples tied everything back to co-sell, which is where many partnership teams feel the urgency most acutely.

Co-sell gets talked about as if it only means hyperscaler motions, but Rob pushed back on that. In practice, co-sell applies more broadly. It is about driving deal velocity with partners and creating shared motion around opportunities.

That kind of motion breaks down quickly when partner teams cannot:

  • explain value clearly
  • package joint messaging fast enough
  • enable internal sellers
  • support partner-facing microsites or assets
  • turn learnings into reusable frameworks

AI helps because it allows smaller teams to create useful support layers without waiting on a long corporate queue.

That does not mean ignoring brand, security, or process. It means recognizing that partner teams have an opportunity to become internal accelerators if they use these tools responsibly.

If partner discovery is one of the bottlenecks you are trying to solve, this playbook on finding best-fit partners faster with AI is another helpful resource.

If you want a practical shortlist from the ideas Justin and Rob explored, start here.

Core AI tools

  • ChatGPT for brainstorming, idea generation, and creative first drafts
  • Claude for structured thinking, long-form drafting, and stronger pushback
  • Gemini for teams already working in Google’s ecosystem
  • NotebookLM for grounded knowledge work based on uploaded source materials
  • Microsoft Copilot for enterprise teams in Microsoft-heavy environments

Workflow and publishing tools

  • Lovable for quickly turning ideas, workflows, and assets into usable interfaces
  • Notion for knowledge organization and databases
  • Google Drive as a central repository or “brain” for your documents and thinking
  • GitHub for saving and managing code repositories
  • Vercel for deployment

How to use this stack wisely

  • Pick one primary model and learn it deeply
  • Keep your source knowledge in a portable, organized location
  • Use a second model occasionally for comparison
  • Optimize for user experience, not just output quality
  • Treat AI as an action layer on top of trusted systems

“I generally used to share spreadsheets. Now I just use Lovable, build the workflow so they can learn how to do it.” -Rob Moyer

FAQs

Which AI tool is best for partnership teams?

There is no single best tool for every partnership team. Rob’s practical answer is to start with the tool your company already approves and supports. Then match the tool to the workflow. ChatGPT may be better for brainstorming, Claude for structured writing, NotebookLM for grounded knowledge work, and Lovable for building shareable workflow experiences.

Should you switch AI tools when a new one becomes popular?

Not automatically. Tools are improving constantly, and they tend to leapfrog each other. Instead of jumping every time the market shifts, master one tool deeply and keep your underlying knowledge in a portable repository like Google Drive or another organized database. That makes switching less painful when you genuinely need to.

How can you avoid losing context when moving from one model to another?

Do not rely on one model’s memory as your only source of context. Store your documents, frameworks, and working knowledge in a central place that you control. Rob’s approach is to create a shared “brain” of source materials, then point different models to that repository so each one can quickly understand the context.

Is prompt engineering still the most important AI skill?

It is still useful, but Rob argues that outcome clarity matters more. Instead of obsessing over perfect prompt wording, explain what you are trying to accomplish and ask the AI to help write the best prompt. That usually produces more useful results and creates a better workflow.

What can partner teams build with AI right now?

Partner teams can build internal knowledge assistants, enablement portals, co-sell frameworks, message libraries, calculators, and training resources. The key is to solve a real workflow problem, not to build something flashy just because AI makes it possible.

Where does Lovable fit into an AI workflow?

Rob uses Lovable as a design and publishing layer. It helps him turn internal knowledge and AI-assisted content into interfaces clients can actually use. For him, that includes decks, workflow guides, and website-like tools that are easier to navigate than static databases or documents.

What is the difference between a system of record and a system of action?

A system of record stores official data, such as your CRM or PRM. A system of action helps you do something with that data, such as creating messaging, generating enablement, or answering partner questions. AI is increasingly becoming that system of action layer for partner teams.

Conclusion

The most useful takeaway here is also the simplest: stop waiting for one perfect AI answer to arrive.

You do not need a final winner between Claude, ChatGPT, Gemini, Copilot, or Lovable. You need a better operating model. Start with your approved environment. Organize your knowledge so it is portable. Match the tool to the task. Compare outputs when it matters. Keep your human judgment in the loop. Then turn what you know into workflows your team and partners can actually use.

That is where the real opportunity is for partnerships right now. Not in chasing novelty, but in building faster systems of action around the work you already know needs to happen.

Partner Playbooks

Your trusted source for B2B SaaS playbooks

© 2026 Partner Playbooks. All rights reserved.