4.7 Sales: Lead Automation & Multi-Source Prospecting
Sales revolves around "clients" and "leads" — collecting business cards at trade shows, following up, developing new retail accounts. Much of the work requires quick response and data support. This chapter shows you how to turn business card photos into leads, auto-research customer backgrounds, intelligently score and rank, and proactively prospect from multiple data sources — so sales teams spend their time closing deals.
The Sales Workflow
Whether you're a sales manager, account manager, or business developer, the daily flow is similar:
| Step | Traditional approach | Pain point |
|---|---|---|
| 1. Trade show lead capture | Collect business cards, organize them at home | Leads go cold by the time you get home, no context for follow-up |
| 2. Customer research | Manually check Google, LinkedIn, Instagram | 4–5 platforms per customer, time-consuming |
| 3. Find new retail accounts | Walk the streets, rely on network, guess | No data on which neighborhoods have target stores |
| 4. Follow-up management | Excel tracking, easy to forget | Leads go cold after a week without contact |
With AOS Work, these four steps happen in one platform: snap a business card photo → instant structured lead, AI auto-researches customer background from multiple sources, Apify proactively searches for target stores, follow-up status is auto-monitored.
Resources Used
| Type | Resource | What it does here | Notes |
|---|---|---|---|
| App | Base | Prospect register, store profiles, follow-up dashboard | Throughout; build the register first |
| Connector | Apify | Connects external sources with actors for Google Maps, LinkedIn, Instagram | Scenarios A and B; public data only |
For adding and authorizing research services, see Connector. Define lead and store profile fields with Base tables and filter or sort follow-up lists with Base views. To query pending follow-ups every day, set the execution time with Scheduled tasks.
Sub-scenario A: Customer Profile & Lead Automation — From Business Card Photo to Follow-up Suggestions
Pain Point
"Collect a stack of business cards at the trade show. By the time I organize them at home, leads are cold. Customer background research is manual. No context for follow-up."
What Makes This Valuable
Snap a business card photo at the trade show → upload → AI parses and writes to the table → auto-researches customer background (Google Maps / LinkedIn / Instagram / website) → AI gives partnership recommendations. From lead capture to follow-up, all in one step. Leads don't go cold.
Steps
-
Snap and upload the business card.
At the trade show, snap a photo and upload it to the AOS Work conversation. -
AI auto-OCR + write to Base
The AI auto-recognizes the card information. -
AI auto-researches customer background.
The AI automatically researches from multiple channels and creates a complete profile:
| Channel | Research content |
|---|---|
| Google Maps | Location, business status, rating, review count |
| Company size, industry, recent updates, key decision-makers | |
| Follower count, partner brands, content style, recent activity | |
| Website | Main business, target customer segment, price range |
- AI gives partnership recommendations.
You input your follow-up rules. The AI gives recommendations based on the customer profile:
| Recommendation dimension | Description |
|---|---|
| Match score | High / Medium / Low |
| Partnership approach | Wholesale / Retail / E-commerce |
| First-contact talking points | Based on the customer's recent activity |
| Recommended products | Based on the customer's segment and price range |
- Set follow-up reminders.
Leads not followed up within 7 days are auto-flagged.
Sample Prompt
1. Google Maps: company location, business status, rating and reviews
2. LinkedIn: company size, industry, recent updates
3. Instagram: follower count, partner brands, content style
4. Official website: main business, target customer segment, price range. Summarize and write to Base customer profile. Based on the customer profile and my follow-up rules (target customer = independent store + mid-high price tier + at least 5000 Instagram followers), provide partnership advancement recommendations:
1. Match score
2. Recommended partnership approach (wholesale / retail / e-commerce)
3. Suggested first-contact talking points (based on customer's recent activities)
4. Recommended products (based on customer's segment and price range) Set reminder: leads not followed up within 7 days auto-flagged in red.
Exercise: Build a customer profile from a card and follow up
Use the business card for the example store Elm Atelier. Select Library and Base to read the company, contact, and email, check for an existing record with that email, and save the lead.
Select the Apify data-collection connector and Base to add store background and score it against the partnership criteria. Record the customer’s catalogue request and next follow-up date for later reference.
Expected Results
-
Business card photo → structured lead
-
AI auto-researches 4 channels, generates a complete customer profile
-
AI gives partnership recommendations based on your rules
-
Follow-up status auto-monitored — leads don't go cold
Advanced: Batch Process All Trade Show Business Cards
Collected 50 business cards at the trade show? Snap all photos, upload in batch. The AI processes: recognize → research → score → rank by match. You start following up with high-match leads by the time you get to the hotel.
Sub-scenario B: Sales Lead Prospecting — Multi-Source Data to Find Stores (Advanced)
Pain Point
"Finding new retail accounts means walking the streets or relying on your network. No data on which neighborhoods have target stores. Customer screening lacks data support."
Steps
- Set the target store profile + search area.
Tell the AI what kind of store you're looking for and where.
| Element | Description | Example |
|---|---|---|
| City/neighborhood | Where to search | London Shoreditch |
| Store type | What kind | Independent fashion boutique |
| Price range | What tier | Mid-to-high end |
| Target customer | Who they serve | 25–35 age group |
-
AI searches Google Maps via Apify.
The AI searches by area + type, returns: store name, address, rating, review count, business status, category. -
AI multi-source deep research.
The AI conducts multi-dimensional research for each candidate store.
| Channel | Research content |
|---|---|
| Google Maps | Location, street view, rating, review content (analyze customer profile) |
| Nearby demographic profile | Population density, spending level, age distribution |
| Company size, founding year, operating history (chain vs. single store) | |
| Follower count, partner brands, content style (brand alignment) | |
| Online published content | Multi-source search for the store's content (selection style, existing suppliers) |
- AI match score + partnership recommendations.
Based on your partnership rules, the AI evaluates each store.
| Recommendation dimension | Description |
|---|---|
| Match score | 0–100, weighted across dimensions |
| Partnership approach | Wholesale / Consignment / Co-brand / Pop-up |
| Strengths | Why this store is a good fit for us |
| Risk alerts | Existing competitor partnership? Mismatched customer base? Unstable operations? |
| First-contact pitch | Based on the store's recent activity and selection style |
- Sort by match score, write to Base.
The AI sorts by match score descending. You start with the Top 5.
Sample Prompt
1. Google Maps: location, street view, rating, review content, analyze in-store customer profile
2. Nearby area demographic profile (population density, consumption level, age distribution)
3. LinkedIn: company size, founding year, operating history (chain vs. single store)
4. Instagram: follower count, partner brands, content style
5. Multi-source search for store's online published content, understand selection style and existing suppliers Summarize and write to Base store profile. Based on the store profile and my partnership rules (target = independent boutique + mid-to-high end + Instagram followers 5000+ + no direct competitor partnership), provide for each candidate store:
1. Match score (0–100 points)
2. Partnership approach (wholesale / consignment / co-brand / pop-up)
3. Strengths analysis
4. Risk alerts
5. Suggested first-contact pitch Sort by match score from highest to lowest.
Exercise: Find stores, score them and prepare first contact
Select the Apify data-collection connector to find fashion stores within two kilometres of Shoreditch in London. Score independent ownership, price tier, Instagram follower count, and competitor partnerships. Select Base to save the profiles and scores to Store Prospects.
Query the two highest-ranked stores, suggest partnership formats and first-contact messages, and mark the example store Elm Atelier as ready to contact.
Expected Results
-
Auto-search for target stores by area — significantly reduce manual street-level research
-
Each store has a profile aggregating the available sources
-
AI scores and sorts by your rules — start with Top 5
-
Includes risk alerts and partnership approach suggestions — first contact has context
Advanced: Batch Search Multiple Neighborhoods
Sample Prompt
Aggregate the rankings and output a city-level partnership opportunity list.
Summary
In short: Agent handles lead research and follow-up reminders; you focus on customer relationships.
4.6 Marketing: Competitor Social Media Monitoring, Content Planning & Campaign Execution
Marketing revolves around "content" and "monitoring" — tracking competitors, creating campaign posters, producing social media videos.
4.8 Merchandise Planning: Product Operations Analysis & Inventory Risk Alerts
One of the core responsibilities for merchandise planning and operations teams is the weekly product analysis — what's selling, what's profitable, what's at risk, what to push and what to clear.