A sales manager exported 12 months of order data for his top 50 accounts. He sorted by product category and noticed something he had not seen in individual transactions: 35 of his 50 accounts bought HRC from him but purchased CRC elsewhere. He was leaving $2.1 million in annual CRC revenue on the table with customers who already trusted him enough to buy HRC. Three months of targeted outreach to those 35 accounts captured $600,000 in annualized CRC business. The data had been in his system for years. Nobody had looked at it this way.
Product Gap Analysis
For each of your top 50 accounts, list the product categories they buy from you and compare it to their total steel consumption (which you can estimate from their industry, size, and conversations with their purchasing team). The gaps between what they buy from you and what they buy total are your cross-sell opportunities.
A machine shop that buys 10,000 pounds per month of flat bar from you probably also buys round bar, plate, and tube from someone. If you stock those products, the cross-sell conversation is simple: "I noticed you buy flat bar from us regularly. We also carry the round bar and plate you use. Can I quote your next order for those products?"
The data makes this conversation specific and credible. You are not cold-calling with a generic pitch. You are approaching a customer you know, referencing their actual buying pattern, and offering a logical extension. Close rates on data-driven cross-sell conversations run 30% to 50%, compared to 5% to 10% for generic prospecting.
Buying Pattern Analysis
Track each customer's ordering frequency and average order size over time. A customer who ordered weekly for two years and has shifted to biweekly is sending a signal. Their total demand probably has not changed. They are splitting volume with another supplier. Catching this shift at week 3 (instead of month 6) lets you intervene while the relationship is still salvageable.
Similarly, a customer whose average order size has been growing steadily is probably growing their business. This is the customer to visit, understand their expansion plans, and position for increased volume before a competitor gets there.
Seasonal and Project Patterns
Some customers have predictable seasonal patterns. A construction-related fabricator buys heavily in March through June and lightly in December through February. If you understand this pattern, you can proactively contact them in February: "Based on your buying pattern, your spring season usually kicks in around the first week of March. Want me to pre-position your typical starter inventory so you are ready to go?"
Project-based buyers show a different pattern: large orders followed by periods of inactivity. Track the average project length and the typical gap between projects. When a project-based customer finishes a large order and goes quiet, estimate when their next project should start and reach out at that time. "Your last project wrapped up about 8 weeks ago. Are you starting anything new that we should be quoting?"
Profitability Segmentation
Combine revenue data with margin data to segment customers into four quadrants: high revenue / high margin (your best accounts, protect them aggressively), high revenue / low margin (volume accounts that may need pricing adjustment or service-level rightsizing), low revenue / high margin (growth opportunities, how can you get more of their spend?), and low revenue / low margin (accounts that may not justify the service cost, consider minimum order policies or reduced service levels).
This segmentation tells your sales team where to invest their time. An hour spent growing a high-margin account is worth more than an hour spent servicing a low-margin account that consumes resources without contributing to profitability. The data makes this allocation objective rather than political.