I'm Ivan. I build AI automation systems. Instead of writing another article about what AI can theoretically do for sales ops, here are three things I actually built last month - with real numbers.
1. Lead enrichment and scoring for a B2B SaaS company
The client was getting 200+ inbound leads per month through their website form. Their sales team spent 2+ hours per day manually researching each lead before deciding whether to reach out.
What we built: when a form is submitted, a workflow fires. Apollo pulls company data. Claude reads the company description and the lead's form copy and assigns an ICP score (1-5) with reasoning. High-scoring leads get routed to a rep with a Slack notification. Low scores go into a nurture sequence automatically.
Results:
- Manual research time: from 2+ hours/day to 20 minutes
- Rep focus: top 20% of leads by score, not random order
- Build time: 6 days
- Build cost: $3,500
- Monthly run cost: $180 (Apollo + Claude API + n8n)
2. Automated pipeline report for a logistics company
The ops manager was spending 3-4 hours every Friday building a pipeline report in Excel for the weekly leadership meeting. Copy-pasting from Pipedrive, calculating conversion rates manually, writing the summary.
What we built: a scheduled job that runs at 7am Friday. Pulls all pipeline data from Pipedrive. Calculates stage conversions, average deal size, deals by rep, week-over-week movement. Claude writes a 6-sentence plain-English summary of what changed. The whole report - data + narrative - is in Slack by 8am.
Results:
- Manual reporting time: from 3-4 hours to 0
- Report quality: more consistent, catches things human summary missed
- Build time: 3 days
- Build cost: $1,800
- Monthly run cost: $60
3. Follow-up sequence triggered by deal activity
The client's reps were forgetting to follow up on deals that went quiet. Deals were sitting in Negotiation stage for 30+ days with no activity. Revenue was leaking.
What we built: a daily job checks every open deal in HubSpot. If there's been no activity (email, call, note) in 7 days, it creates a task for the rep with context: last interaction, deal size, days in stage. If 14 days with no activity: the rep's manager gets a Slack alert. The rep can't forget - the system won't let them.
Results:
- Deals with 30+ day no-activity: dropped from 40% to 8% of pipeline
- Reps found the alerts more useful than annoying
- Build time: 2 days
- Build cost: $1,200
- Monthly run cost: $30
Total investment across three automations: $6,500 build, $270/month running. Total manual time replaced: roughly 50 hours per month across the teams. At $60/hour fully loaded, that's $3,000/month in labor replaced. Payback period: about 2 months.
If you want to see what the equivalent would cost and save for your sales ops specifically, the 4-question audit at 2pizza.team/audit takes 2 minutes and gives you a personalized breakdown.
Free 30-min audit. We tell you what to automate first and what it would cost.