1 of 39

Scaling Account Research with No-Code AI

1

ZAPIER

2 of 39

2

About Us

Blake Cohlan

Sr. Director, Growth Marketing

SupportLogic

Reid Robinson

Lead AI Product Manager

Zapier

ZAPIER

3 of 39

Live Demo: Scaling Account Research

Workflow Breakdown

Starter Workflows for Marketers

01

02

03

Templates, Tools & Community

04

Open Build & Q&A

05

AGENDA

3

ZAPIER

4 of 39

Why Now, Why This

  • Account research is the fuel for targeted, high-converting plays across marketing, sales, and CS

  • We're going to show you how to:
    • Automate research and save 5+ hours/week per rep
    • Build scalable workflows without needing a degree in coding
    • Replace one-off plays with repeatable, measurable programs

4

ZAPIER

5 of 39

5

1:1 Account Planning Template

SDR Resources

Marketing Resources

Pre-Sales Support

Prospect Analysis

Specific outbound actions, such as personalized email sequences, LinkedIn outreach, or direct mail campaigns.

Specific personalized requests to Marketing (eg. Send book to stakeholders, create a virtual or in person innovation day, drip series of emails around product or whitespace)

Specify requests for solution engineers or consultants to address technical objections or assist in proof-of-concept stages

Prospect Business Pains:�Expected challenges based on discovery or research.���Current Process or Solution:�Current technology or process, including gaps we fill.���Competitive Intelligence: �Using direct or adjacent solutions? Identify renewal dates or any dissatisfaction.��

Expansion Potential: What’s the long-term potential for account expansion once they’re a customer.

Revenue Forecast

Prospective Buying Committee

List other influential people in the organization.

Time to Close estimate?

Quarter forecasted to close? Sales Cycle Length

Estimated potential revenue based on deal size, contract length, and upsell potential.

Product Resources

Specific requests for the product team.

Prospect Engagement Plan

How will you engage and progress the account?

Org Pain Points and Priorities

Challenges or strategic priorities faced. How does SL Support?

Estimated Budget Allocation

Potential Revenue Opportunity

Estimate budget needed based on benchmarks or current tech investments.

What is estimated ARR?

Key Decision Makers

Include names, roles, and influence of key executives

ZAPIER

6 of 39

Demo

6

ZAPIER

7 of 39

7

ZAPIER

8 of 39

8

ZAPIER

9 of 39

9

ZAPIER

10 of 39

10

ZAPIER

11 of 39

11

ZAPIER

12 of 39

Testing

13 of 39

Testing

14 of 39

Opportunity Analysis

  1. SFDC - Get Opps
  2. SFDC - Query Opps
  3. OpenAI - Summarize
  4. Format (Rich Text)
  5. Salesforce Writeback

14

ZAPIER

15 of 39

Breakdown

  • Trigger & Setup
  • External Research (Perplexity)
  • AI Assistant (OpenAI)
  • Formatting & Personalization
  • Salesforce Writeback

15

ZAPIER

16 of 39

Button click kicks off a zap pulling data from Salesforce and running an analysis through Perplexity and OpenAI

ZAPIER

17 of 39

API into sec.gov for 10k and earnings call analysis

ZAPIER

18 of 39

“Trained” OpenAI assistant runs analysis and generates cross-functional summaries.

ZAPIER

19 of 39

Trigger and Setup

  • Zapier will provide you with a webhook URL

19

ZAPIER

20 of 39

Trigger and Setup

  • New Button or Link on Buttons, Links, and Actions, under Account Object settings.

20

ZAPIER

21 of 39

External Research (Perplexity)

  • Set up API at Perplexity and add credits
    • $20 to start
    • Don’t need pro account�
  • Research:
    • Company Overview + Market Signals
    • C-Suite Research
    • Revenue & Retention Challenges
    • Job Listings & Support Roles
    • How does the company talk internally about their products and solutions?

21

ZAPIER

22 of 39

External Research - Lessons Learned

  • Basic Sonar Model ($.01 per call) is good enough - Test for your use case

  • OpenAI 4.0 outperformed Sonar-Pro
    • Relies on indexed data; If executives don’t frequently appear in external news articles, blogs, or media, they might not be indexed.�
  • Easy to burn $$ on testing different models or revising code with errors
    • Deep research cost me $0.60 for one run and couldn’t write back because of size error.�
  • Troubleshoot and revise code with OpenAI (screenshots and questions) ie. Why am I getting this error? How can I improve this? What does this mean in the API settings?

22

ZAPIER

23 of 39

Summarization & Synthesis (OpenAI)

  • Assistant Type: Custom GPT-4 Assistant

  • Training Source: Internal docs (persona briefs, positioning, case studies)

  • Function: Synthesizes all raw research into structured insight
    • Growth focus areas
    • Key pain points
    • Strategic alignment with your company

23

ZAPIER

24 of 39

Summarization & Synthesis (OpenAI)

24

ZAPIER

25 of 39

Salesforce Writeback

Automatically updates:

  • Latest Company Summary
  • Engagement Ideas
  • Persona Messaging
  • Campaign Concepts

25

ZAPIER

26 of 39

Can you do more with less?

26

Manual Research

Automated Workflow

Time per Account

~45–60 min

~60–90 sec

Tech (per run)

~$12 (based on blended rate)

~$0.83 (Zapier + OpenAI + Perplexity)

Output Quality

Inconsistent, varies by rep

Structured, tailored, always-on

Scalability

5 accounts/week/rep

50+ accounts/week, repeatable

Personalization

Spray and Pray

Deeply tailored, prompt-driven

Messaging Alignment

Depends on rep skill

Auto-aligned to personas, value prop

ZAPIER

27 of 39

Summarize content

Write content

Reply to customers

Extract data

Classify data

Top AI

Uses

28 of 39

Trigger

Actions

29 of 39

So…How does all

this work on Zapier? 🧐

29

ZAPIER

30 of 39

30

ZAPIER

31 of 39

Chatbots

32 of 39

Agents

33 of 39

Start with the OpenAI API

  • Access the same models powering ChatGPT�
  • Build AI Assistants similar to GPTs�
  • Test your prompts in the Playground�
  • And of course, generate your API Key 🔑

33

ZAPIER

34 of 39

Levers to Use

34

User Message

Your message to ChatGPT

Assistant

Instructions

Background details

Model

GPT-4o Mini and GPT-4o are both great

Memory Key

A unique value allows you to continue the conversation from another Zap run

Temperature

Controls randomness. Lower to consistent results similar inputs

Max Tokens

Increase for longer responses

ZAPIER

35 of 39

Get Familiar with Prompting

Things to Remember:

  • Use dynamic variables�
  • Provide relevant context�
  • Use Custom Fields�
  • Test prompts in the OpenAI Playground

35

ZAPIER

36 of 39

Exclusive for Workshop Attendees

Take Zapier’s new AI Readiness Quiz, a quick way to see where you stand and what to do next.��✅ Quick quiz

🎁 Tailored AI insights

🚀 Real recommendations for what to do next

Afterwards you will get a personalized score and resources to help move forward and some folks may also be invited to a 1:1 strategy session with Phil, Zapier’s Head of AI Innovation.�

Unlock a free session with Zapier’s AI lead, Phil.

36

ZAPIER

37 of 39

Workshop

Time! 👨‍💻

37

ZAPIER

38 of 39

Q&A

ZAPIER

39 of 39

Thank you

zapier.com/ai

linkedin.com/in/reidtrobinson

linkedin.com/in/blakecohlan

39

ZAPIER