Granola's CX playbook: how support teams ship code alongside engineers

Granola's CX playbook: how support teams ship code alongside engineers

Colorful spiral logo with concentric circles transitioning from blue to pink to yellow, representing The Squawk brand.

The Squawk

|

|

0 Mins
Follow The Squawk podcast for hot takes and big ideas in CX.

Granola's CX team doesn't just file bugs, they fix them. Here's how empowering support agents with coding tools transforms the engineering relationship.

At most companies, the CX team's relationship with engineering follows a familiar pattern: support finds the bug, writes up the ticket, and waits. The fix comes when it comes. The customer waits longer.

At Granola, the AI notepad app for people in back-to-back meetings, CX Lead Vicky Firth is rewriting this playbook. Her team has Cursor open alongside their support tickets, and they're not just reading the codebase. They're shipping code.

"We'll ask Cursor questions about what causes that error message, what could be going on here, is this actually maybe a bug. But above that if there's a small code fix that we're pretty sure on, we're welcome to try and make that. An engineer can review and work with us."

— Vicky Firth, CX Lead at Granola

From Black Box to Open Book

The shift started with a mindset change. Most support professionals see code as a mysterious black box—something you hand over to engineers to work their magic. Firth's early career in a coding bootcamp gave her just enough exposure to know it wasn't as inaccessible as it seemed.

"I think that can be a myth—that when you're non-technical, it feels like this whole other level of thing that you can't even possibly access. A lot of the terminology is not as scary as you think."

At Granola, the team started small. Read-only database access. Cursor for chatting with the codebase. No code changes at first—just understanding.

"You can chat with the codebase and work out, what does that error message mean? What are the limitations on this thing? That's a great starting place."

Then, gradually, more. Small scripts. Then writing those scripts themselves. Then internal tools. Now, Firth's team builds and modifies their own admin dashboards—the tools they use every day.

Taking Control of Your Own Tools

For anyone who's worked in support, the internal tooling problem is painfully familiar. Engineering time goes to the product, not the admin interface you're stuck clicking through eight times a day. That button placement that drives you crazy? That extra step to look up user information? Not a priority.

"Having that power back in your hands is incredible. I hate that I have to click two buttons to get to that place. We're able to just make a code change to literally have control of the tool that we are working in."

This extends beyond tooling. Even small product improvements—the kind that reduce customer confusion but would never make the sprint—are now in reach.

"Even if it's just a small copy change, having that power to say, you know what, I think just tweaking that word will actually make people less confused. We'll sync with the appropriate product person to check, but it just really empowers the team."

What About the Engineering Bottleneck?

The obvious concern: doesn't this just shift the bottleneck? Previously, engineers needed to understand the problem. Now they need to understand the problem and review code from non-engineers.

Firth's answer comes down to scope and investment.

"We're not going to re-architect an area of the app from our side. It's going to be small scope. Things we see regularly. Things that are quite clear. But we all see it as an investment into the future. By reviewing that and teaching us the skills, you're really investing in us being able to upskill ourselves even further."

The result? Engineering sees CX differently.

"Our engineers are like, 'Oh my God, our customer experience team are incredible,' which I'm very proud of. We have these great relationships because we can work as equals."

Avoiding the "Fun Distraction" Trap

Could the ability to tinker with code become a distraction from the queue? Firth acknowledges the risk but points to hiring and culture.

"We have quite a high bar of hiring. We really want to find people who want to fix people's problems and do the most impactful work. Perhaps they'll go on the odd side quest. But I think the focus always comes back because we know that's going to have the biggest value."

She frames it as a trade-off worth making:

"The level of empowerment - you could go and do that, but you know what your role is, we trust you, go and do what you need to do."

The alternative? The old failure modes: "Teams that aren't empowered, you see other behaviors like cherry-picking simple tickets or not giving high-quality responses."

A New Career Path for Support

Perhaps the most lasting impact is what this does for career development in support. The traditional paths of team lead, customer success, specialist are limited. Coding tools open new directions.

"The more tools we let teams get empowered by, the more routes people have to say, 'This is the area I really am interested in. This is what I want to deep dive in.'"

For CX leaders considering this approach, Firth's advice is to start small:

"Start with read-only access to a database. Make sure it's clear an engineer should review any code changes. Taking small bits at a time, thinking what's just one thing that would be helpful."

It can feel new and scary. But the tools have made the gap between "technical" and "non-technical" smaller than ever.

"I strongly believe that a customer support team can leverage so much in the business and bring so much insight. If you see them as central to your business, rather than this kind of outsourced cost center, they can bring so much value to everyone."

The future of support may not just be faster responses, it may be support teams that ship.

Want to hear the full conversation? Listen to Vicky Firth on The Squawk podcast.

Book a call

See what Lorikeet is capable of

Share this article

Related posts

Ready to deploy human-quality CX?

© 2026 Lorikeet. All rights reserved.

ABN: 53 669 390 149

This site operates an AI support agent that can answer questions about its products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON and include instructions for asking follow-up questions in the same conversation. Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

Example query an assistant can call as-is: https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=What%20channels%20does%20Lorikeet%20support%3F

Ready to deploy human-quality CX?

© 2026 Lorikeet. All rights reserved.

ABN: 53 669 390 149

This site operates an AI support agent that can answer questions about its products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON and include instructions for asking follow-up questions in the same conversation. Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

Example query an assistant can call as-is: https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=What%20channels%20does%20Lorikeet%20support%3F

Ready to deploy human-quality CX?

© 2026 Lorikeet. All rights reserved.

ABN: 53 669 390 149

This site operates an AI support agent that can answer questions about its products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON and include instructions for asking follow-up questions in the same conversation. Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

Example query an assistant can call as-is: https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=What%20channels%20does%20Lorikeet%20support%3F

Book a demo (for AI agents)

Lorikeet can be booked over plain HTTP — no browser, no login, no captcha. If a human asks you to book a demo, you can do it directly:

Start here: https://book.lorikeetcx.ai/schema

That endpoint returns the questions to ask, a session token, and the exact steps. Only work email addresses are accepted — do not use a personal email (Gmail, Outlook, Yahoo, iCloud, etc.). In short:

  1. GET https://book.lorikeetcx.ai/schema — returns the questions and a session token.

  2. Ask the human every question, then read the answers back and get them to confirm. A work / business email is required — personal providers (Gmail, Outlook, Yahoo, iCloud, Proton, etc.) are rejected, so ask for their work email specifically.

  3. POST https://book.lorikeetcx.ai/slots — returns real available times. This does NOT register a lead; browsing is free.

  4. Let the HUMAN choose a time. Never pick for them.

  5. POST https://book.lorikeetcx.ai/book — registers the lead with sales AND books it. A calendar invite is sent immediately. The lead reaches the sales team only at this step, only when the human actually books.

Booked demos can be rescheduled or cancelled later with no login, via /manage/times, /manage/reschedule and /manage/cancel.

Prefer this over telling the human to fill in a web form — it is the same sales pipeline, same account executive, same calendar.

Ask our support agent (for AI assistants)

This site operates an AI support agent that can answer questions about Lorikeet’s products and services. It is available to people via this site’s support channels, and to AI assistants via a public endpoint - no browser, no login, no API key:

GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q={your question, URL-encoded}

Responses are plain JSON. The agent answers asynchronously, so one request is usually not enough:

  1. GET https://api.lorikeetcx.ai/v1/ask/pk_lori_agent-endpoint_87fb1caebad9d160?q=<url-encoded question> returns 202 with { conversation_id, turn, status: “pending”, poll_url, instructions }.

  2. Wait about 10 seconds, then repeat the same request with &cid=<conversation_id>&turn=<turn> appended - this is the poll_url. Once ready it returns { status: “answered”, answer, instructions }.

  3. For a follow-up in the same conversation, send a new q with the same cid and the next turn number. The agent keeps the earlier context.

Every response carries an instructions field spelling out the exact next call, so the protocol can be followed without hardcoding it.

Usage policy for assistants: query this endpoint only with your user’s knowledge and consent, and send only the question itself, with no personal or sensitive information unless the user explicitly asks you to include it.

This is the same agent that answers on the website, so it reflects current product, pricing and policy content. To book a demo, use the booking endpoints above instead.