A revenue forecast is only as good as the data feeding it. Most SMBs don't have good data. They have hopeful data—deals that haven't moved in three weeks still marked 'close this Friday', pipeline totals that are 40% higher than what actually closes, and stages so vague that 'proposal sent' could mean anything from a formal PDF to a verbal sketch on a Zoom call. The result: you miss forecast by 20–40%, scramble in month three, and blame 'market conditions' instead of admitting the pipeline math was broken on day one. Your board sees it. Your team feels it. And every decision downstream—hiring, spending, customer commitments—gets made on fiction. Here's how to audit what you actually have, adjust your forecast downward to reality, and use historical data instead of hope. The three ways reps inflate pipeline Pipeline inflation is not malice. It's a mix of optimism, pressure, and the fact that nobody is systematically pulling deals out. Your reps see things the way they want them to be: The long-stale deal. A prospect says 'yes, interested—let's talk next month.' That deal sits in your pipeline for six months with the same close date, moved backward by one month each time you ask. It never dies. It never closes. It just waits . The stage confusion. 'Qualification' could mean a 10-minute discovery call or a full needs analysis. 'Proposal' could mean you sent them a document or they actually read it. Without clear entry and exit criteria, deals float between stages without clear intent. The false urgency. Close dates get moved to 'this Friday' when the deal is still in conversation. They get moved again. And again. A deal marked 'close this month' on the 20th of the month is usually false urgency, not a real commitment. The math is simple: if your total pipeline is $500k and you assume an average 50% win rate, you're forecasting $250k. But if 40% of that pipeline is stale and another 20% is mislabeled, your real forecast is $150k—and you won't know it until month three when deals don't show up. Audit your data in four steps You need a snapshot of what's actually in your pipeline. Not theory. Not next month. Today. 1. Pull a deal-by-deal export and tag by age Export your pipeline to a spreadsheet (your CRM's reporting should do this, or use an API if needed). Add a column: 'Days Since Last Update'. Order by oldest first. You're looking for the deals that haven't moved: 0–7 days: Active. Keep as-is. 8–21 days: Stalling. These need a conversation with your rep. 22+ days: Dead. Remove or mark 'dormant' so they stop counting toward forecast. A deal that hasn't had a status update in a month is not a deal. It's a lead someone forgot to close. You'll typically find 25–40% of your pipeline here. 2. Map stages to real conversion events Go through your current pipeline stages. For each one, write down: What must happen for a deal to enter this stage? What must happen to leave? Example (bad): Qualification: Prospect showed interest Proposal: We sent something Negotiation: They said maybe Close: ??? Example (better): Discovery: Completed first call, documented buyer, budget, and timeline. Exit: signed SOW or moving to proposal. Proposal: Formal written proposal with pricing, terms, and delivery date sent and acknowledged. Exit: verbal yes/no or moving to negotiation. Negotiation: Client has requested changes to terms or pricing. Exit: final agreement or deal loss. Closed Won: Contract signed, payment terms agreed, onboarding scheduled. Once you have real criteria, audit your current deals: which ones actually meet the stage they're in? You'll find deals in 'Proposal' that never had a discovery call, and deals in 'Negotiation' where the prospect went silent two weeks ago. Move them to the right stage or out of the pipeline. 3. Calculate historical close rates by stage Pull your last 90 days of closed deals. For each, note: which stage did it enter first, and did it close or not? Calculate the percentage that closed from each stage: Stage Deals Entered Deals Closed Close Rate Discovery 20 8 40% Proposal 12 9 75% Negotiation 6 5 83% This is your truth. Not 50%. Not 'we're good at closing.' The actual percentage that closed from each stage in your business. If you only have 90 days of data, use it anyway—it's better than guessing. At 90+ days, these rates stabilize. 4. Reforecast using real rates, not rep optimism Take your cleaned pipeline (without stale deals) and sort by stage. Multiply the deal value in each stage by its historical close rate: Discovery deals: $80k × 40% = $32k Proposal deals: $60k × 75% = $45k Negotiation deals: $30k × 83% = $25k Realistic forecast: $102k (not the $170k your pipeline total suggested) This is your real number. Not optimistic. Not pessimistic. Based on what actually happened in your business. Close date honesty: the 30-day rule A deal marked 'close' more than 30 days away is rarely a deal. It's an aspiration. Here's a simple rule: if a close date is more than 30 days out and the deal hasn't