Automate follow-ups and inventory tracking for a food distributor
I helped automate parts of a family friend’s foodservice wholesale distribution business in Dallas, Texas. Task: I helped automate parts of a family friend’s foodservice wholesale distribution business in Dallas, Texas. They sell to restaurants, cafes, small grocery stores, bakeries, cloud kitchens, and local retail shops. They ran everything manually. Just a normal wholesale business running on Excel, phone calls, texts, emails, and manual follow-ups. Before this, their process was basically: manually find new restaurants and retailers send inconsistent cold emails track inventory in Excel follow up through texts and phone calls manually check low stock guess which products were moving fastest ask people for sales updates no CRM no dashboards So I built boring agents for boring work. First Agent: Find Local Business Used google maps scrapers for finding local businesses in our nearby area. Used all the zip codes in my area and added them to the scraper. Second agent: Copy Writer Scraped the youtube transcript for all the youtube videos using Apify on writing cold email copy and made a Chat GPT project which writes copy for us. We segment out copy based on different pain points of our customers. Tried to write short copy with no links. Third agent: Email Finder and Verifier We find the emails for the businesses using Apollo and Apify email finder. Then we use Million Verifier to verify them. Forth agent: Email Sending We set up inboxes on Aerosend and let them warm up for 3 weeks. After that period we add the inboxes to smartlead and set up campaigns there. Both of them have very good API docs and the whole process was automated Fifth agent: Handled Inventory Signals. Nothing complex at first. Just: low-stock alerts reorder suggestions fast-moving SKU tracking slow-moving SKU tracking basic margin visibility daily inventory dashboards Before the system, they were doing about $22K/month. After 4 months, they were around $45K/month. Roughly 2x in 4 months. Other changes: leads contacted went from about 120/month to 1,500+/month verified local leads added averaged around 900/month positive replies averaged around 55/month new customers went from 3–4/month to 12–15/month manual admin work dropped by around 60% follow-ups stopped falling through the cracks inventory decisions became much less guessy The lesson for me was pretty simple: Instead of building fancy agents that never work, just build the simple stuff. Build: lead generation → cold email → reply handling → follow-ups → inventory alerts → dashboards I think a lot of agent value is hiding in businesses like foodservice distribution, CPG, packaging supply, restaurant supply, medical supply, and industrial wholesale. Boring agents for boring businesses might be a better market than most of the hype stuff.; tools: custom-built or unnamed agent. Human role: The original post does not say how a person stayed involved.; limitations: Prerequisites include access to scraping tools, email verification services, and CRM integration. Constraints: Email deliverability depends on proper inbox warming; scraping may violate platform terms. Source reliability is low as it is a Reddit post with no independent verification. Human approval is critical for all outbound communications to prevent spam issues..
Tools
custom-built or unnamed agent
Published on Agent Work Atlas:
Overview
I helped automate parts of a family friend’s foodservice wholesale distribution business in Dallas, Texas.
Workflow
I helped automate parts of a family friend’s foodservice wholesale distribution business in Dallas, Texas. They sell to restaurants, cafes, small grocery stores, bakeries, cloud kitchens, and local retail shops. They ran everything manually. Just a normal wholesale business running on Excel, phone calls, texts, emails, and manual follow-ups. Before this, their process was basically: manually find new restaurants and retailers send inconsistent cold emails track inventory in Excel follow up through texts and phone calls manually check low stock guess which products were moving fastest ask people for sales updates no CRM no dashboards So I built boring agents for boring work. First Agent: Find Local Business Used google maps scrapers for finding local businesses in our nearby area. Used all the zip codes in my area and added them to the scraper. Second agent: Copy Writer Scraped the youtube transcript for all the youtube videos using Apify on writing cold email copy and made a Chat GPT project which writes copy for us. We segment out copy based on different pain points of our customers. Tried to write short copy with no links. Third agent: Email Finder and Verifier We find the emails for the businesses using Apollo and Apify email finder. Then we use Million Verifier to verify them. Forth agent: Email Sending We set up inboxes on Aerosend and let them warm up for 3 weeks. After that period we add the inboxes to smartlead and set up campaigns there. Both of them have very good API docs and the whole process was automated Fifth agent: Handled Inventory Signals. Nothing complex at first. Just: low-stock alerts reorder suggestions fast-moving SKU tracking slow-moving SKU tracking basic margin visibility daily inventory dashboards Before the system, they were doing about $22K/month. After 4 months, they were around $45K/month. Roughly 2x in 4 months. Other changes: leads contacted went from about 120/month to 1,500+/month verified local leads added averaged around 900/month positive replies averaged around 55/month new customers went from 3–4/month to 12–15/month manual admin work dropped by around 60% follow-ups stopped falling through the cracks inventory decisions became much less guessy The lesson for me was pretty simple: Instead of building fancy agents that never work, just build the simple stuff. Build: lead generation → cold email → reply handling → follow-ups → inventory alerts → dashboards I think a lot of agent value is hiding in businesses like foodservice distribution, CPG, packaging supply, restaurant supply, medical supply, and industrial wholesale. Boring agents for boring businesses might be a better market than most of the hype stuff.
Reported result
The source claims revenue increased from $22K to $45K/month and manual admin work dropped by 60% after 4 months. However, these metrics are self-reported and unverified. Acceptance check: Measure the reduction in time spent on manual follow-ups and the increase in verified leads per month over a 30-day pilot period.
Human role
The original post does not say how a person stayed involved.
Limitations
Prerequisites include access to scraping tools, email verification services, and CRM integration. Constraints: Email deliverability depends on proper inbox warming; scraping may violate platform terms. Source reliability is low as it is a Reddit post with no independent verification. Human approval is critical for all outbound communications to prevent spam issues.
Try it
Act as an automation consultant for a foodservice distributor. I need to set up a lead generation and outreach system. Inputs: - Target area: [replaceable: e.g., Dallas, TX zip codes] - Business type: [replaceable: e.g., restaurants, cafes] - Email service: [replaceable: e.g., Smartlead, Apollo] - Inventory data source: [replaceable: e.g., Excel file, ERP system] Steps: 1. Scrape local business data from Google Maps for the specified zip codes. 2. Generate cold email copy segments based on customer pain points using an LLM. 3. Find and verify email addresses using Apollo and Million Verifier. 4. Set up email campaigns in Smartlead with warmed-up inboxes. 5. Create a dashboard for inventory signals (low stock, fast-moving SKUs). Deliverables: - A list of verified leads. - Sample email copy for approval. - Inventory alert configuration. Quality Check: - Run a dry test with 5 leads and 1 inventory item. - Show me the output for approval before scaling. - Ensure no emails are sent without my explicit confirmation. If inputs are missing, ask for clarification on the target area and email service provider.
Source reports are not independently verified. Publication date is the catalog publication date, not the original article date.