Build an AI Custom Automation - Sales to CS Handover

Dive into this hands-on walkthrough to build an AI-powered "Sales to CS Handover" automation in Planhat. Step by step, you’ll learn to define your outcome and decide data inputs, set up the trigger, design the AI step and set up actions— so your Sales team can save time and your CSMs can hit the ground running with clarity and context.

24 seconds

Planhat

This guide is designed for you to follow along and build automations step-by-step. As we go through each section, we recommend creating the setup in your own Planhat environment at the same time. Each organisation will have its own specific needs, so use the steps as inspiration to create an automation that works for your business.

Summary: Build an AI Sales to CS handover automation

**Step 1: Decide the outcome and goal** - Outcome: Saved time for the AE and seamless, insight-driven transition where CSMs instantly understand customer context, goals, and next steps — leading to faster time to value and stronger retention - Goal: Leverage AI to eliminate manual handovers, ensure consistency, and deliver a smooth, high-quality customer experience **Step 2: Identify the inputs** - Inputs: End User details, recent Conversations and Opportunity data — so the AI can accurately summarize the customer story and highlight what the CSM needs to know for the onboarding - Trigger: When the company phase updates **Step 3: Design the AI step** - Use Planhat's Managed LLM and GPT-4 - Build a prompt that acts and reasons as a strategic analyst **Step 4: Decide the actions** - Output: Create a handover note or Document on the Company profile **Step 5: Test and Iterate** - Test your newly built automation until you are happy with the output

Step 1: Decide the outcome and goal

The transition from Sales to Customer Success is a critical moment where context often gets lost — what was promised, customer goals, or potential risks. This can cause delays, confusion and a poor customer experience. An AI-powered handover ensures no information slips through the cracks. It automatically summarizes key deal details, customer objectives, and next steps, giving the CSM instant clarity without relying on manual notes or meetings. **Outcome:** A seamless transition where the AE saves time and the CSM instantly understands the customer’s context, goals, and next steps. Leading to faster time to value, stronger relationships, higher adoption, and improved retention. \ \ **Goal:** Leverage AI to save time by eliminating manual handovers, ensure consistency, and deliver a better customer experience.

Step 2: Identify the inputs

Now that you know what you want to achieve, consider **what data the AI needs**. The goal is to give the AI enough context to summarize the customer story accurately and highlight what the CSM needs to know next. For this use case, good inputs could be: - **End user details:** Who the customer is, what they’re trying to achieve and their role or use case - **Conversations:** The most recent and relevant sales interactions to capture tone, sentiment and expectations - **Opportunity data:** What was sold, pricing or terms and any key promises or success criteria These inputs would provide the AI with a comprehensive picture — ensuring the handover summary reflects not just **what** was sold, but **why it matters** to the customer.

When it comes to **triggering** the automation, a natural time to kick it off is when the AE closes the deal in Planhat. You can have this **triggered by an update to a Company field.** Note that this will vary depending on your setup in Planhat. For example, suppose the AE uses a Board of all opportunities. When the deal is sealed, they move the Opportunity to “Deal closed". That triggers an automation that updates the Company’s Phase to “Onboarding,” which in turn triggers the “Sales to CS Handover” automation.

You could also choose to set it up as a **manual trigger** on the Company profile. For instance, if you’ve launched a Sales to CS workflow with defined next steps, ahead of a scheduled handover meeting, the AE or CSM can manually trigger the automation to generate the handover notes.

Next up is to configure the Get steps! As earlier determined for this one we want to: - Get Conversations - Get End Users - Get Opportunities

In each step you want to configure the following settings and filters for all steps: **Object: Multiple** This collects multiple objects (conversations, end users, opportunities), as we want to get a summary of everything in the account, not just a single object. **Filter: Match all - Company ID Equal To <<object.\_id>>** This filter ensures that you only collect data from the specific company you are running the automation on. Additionally, for the Get Opportunity step, you also want to set up: **Filter: Sales Stage Equal To Closed Won** These fields might have different names in your app, so please tweak it so it makes sense for your business. The outcome should be only to get the closed opportunities, as there might be other still open that are not relevant for this handover.

Step 3: Design the AI step

Time for the star of the show - the AI step! In this example, we’re using **Planhat’s Managed LLM**, but you can also select your own model from the **“Provider”** dropdown if you’ve connected an external LLM. The AI model choice is up to you. For tasks like summarizing long sales notes or customer conversations, models such as **ChatGPT**, **Claude**, or **Gemini** all perform well. We recommend experimenting with a few models to see which one delivers the most accurate and relevant summaries for your handover notes.

Now to the **prompt**! There are many ways to write your prompt, but, as mentioned before, the more instructions, the better! It's worth spending time testing different prompts until you are satisfied with the output you get. Treat the AI as a new colleague that needs all context to do its job - the more information you can give, the better. Below you can see an example of the prompt we have done for this specific use case as inspiration: \ \ **Role and Objective**: Act as a strategic analyst reviewing CRM data to extract key business insights about the company <<[object.name](http://object.name)>> for onboarding and success planning. **Data Set:**\ Company: <<object>>\ Conversations: <<NAME OF STEP>>\ Endusers: <<NAME OF STEP>>\ Opportunities:<<NAME OF STEP>> **Reasoning Steps:**\ Parse and synthesize insights from conversations and enduser data.\ Identify recurring themes, strategic language, and stakeholder roles.\ Match observed data to known Customer Success models and business goals.\ Distinguish between mentioned use cases, timelines, tools, and metrics. **Instructions:**\ Only respond where data is present; otherwise reply “not enough data.”\ Prioritize accuracy and structure.\ Interpret only what is explicitly or strongly implied by the data. **Output Format:**\ Provide responses to each of the following questions in numbered format (1–10). Use clear, concise sentences and organize subpoints where needed. Return only plain-text, no bolded formats. **Questions:** 1. What is their business model and how do they generate revenue? 2. Who are the key people and what are their roles or objectives? 3. How do they approach Customer Success (e.g., tech-touch, high-touch, structure, team roles)? 4. What are their high-level business objectives (e.g., net retention, time to value)? What challenges do they face as a business? 5. What are their intended Planhat use cases and how do they plan to measure success (metrics)? 6. Which use cases are prioritized, and what are their expected launch dates or time-to-value expectations? 7. Are there other specific milestones or goals identified? 8. What key tools or platforms do they plan to integrate during onboarding? 9. Where are their primary contacts located? 10. Is there a Service Level Agreement (SLA)? If so, specify terms (only relevant for large accounts). **Context:**\ This prompt is intended to summarize onboarding-relevant insights for the company <<[object.name](http://object.name)>> using internal data available in Planhat. **Final Instructions:**\ Do not make assumptions. Clearly flag missing information. Aim to support Customer Success and onboarding planning with structured, data-driven insights.

Step 4: Decide the actions

Now that you have your output, it's time to decide where you want it to end up. You could choose to **create** a Conversation, **create** a Document or **send** it directly to the CSM - the options are down to you! In our example, we have decided to create a conversation of type "Handover" on the Company profile. The minimum fields you would need to fill in are: - **Type:** Type of Conversation - **Subject**: Title of the Conversation - **Company Id:[[ <<object.\\\\\\\\\_id>>]]** - **Description**: [[<<name of AI step>>]]

If you are creating a Document you need to fill in following fields: - **Owner** - **Name** - **Description:** <<name of AI step>>

Step 5. Test and iterate

So there you have your AI automation! Now it's time to put it to the test. We recommend testing the automation on a Test Company to ensure it behaves as expected. Trigger the automation, check the output, tweak the prompt, and run it again until you are satisfied with the result.

Here you can see it in action!

You’ve now seen how simple it is to turn AI into your handover sidekick — saving time and delivering smoother transitions from Sales to CS. 🎯\ \ Ready to put it into action? Jump into your own Planhat workspace and **build your AI-powered handover today!** Don't forget to check out our other guides with more custom Automations in our How-to-library, or enroll in the "AI step in Automations" course in the Dojo.