Sales prediction does not tell you what will happen. It tells you which deals look most like the ones you have won before<\/strong>.<\/p>\n That is a narrower claim than the marketing, and it is still useful. A salesperson with forty open opportunities and time for fifteen calls has a real prioritisation problem. Ranking them by resemblance to past wins is a better starting point than working down the list by date.<\/p>\n The output is an ordering, not a forecast.<\/strong> Treat it that way and it helps. Treat it as a prediction and it will disappoint you.<\/p>\n The score comes from your own history \u2014 the deals you closed and the deals you lost.<\/p>\n It looks at patterns across things like:<\/p>\n None of this is magic. It is noticing that deals over a certain size from a certain source with no contact for three weeks tend not to close.<\/p>\n FIGURE 1: WHERE A SCORE COMES FROM<\/p>\n Your closed deals<\/p>\n Patterns found<\/p>\n Open deals scored<\/p>\n A salesperson decides<\/p>\n You need enough history.<\/strong><\/p>\n This is the single thing that determines whether sales prediction is useful or decorative in your business.<\/p>\n A few hundred closed deals is thin. Several thousand, across a couple of years, gives something to learn from. If you close twenty deals a year, a score is a guess with a confident interface<\/strong> \u2014 and it will look exactly as authoritative as a good one.<\/p>\n Two other data requirements matter as much:<\/p>\n You must record losses properly.<\/strong> A model trained only on wins cannot tell you what a loss looks like. If your team marks deals lost without a reason, or leaves them open forever, half the training data is missing.<\/p>\n Your pipeline must reflect reality.<\/strong> If deals sit in “Negotiation” for eight months because nobody moved them, the model learns that stage means nothing.<\/p>\n Three uses that work.<\/p>\n Ordering the day.<\/strong> Which five opportunities deserve a call this morning. This is the main one and it is worth having.<\/p>\n Spotting stalled deals.<\/strong> Opportunities whose score has dropped \u2014 activity has fallen off, or they have sat too long. These are usually deals people have quietly given up on without saying so.<\/p>\n Sanity-checking the forecast.<\/strong> When a salesperson says a deal will close this month and the score disagrees, that is a conversation worth having. Not an argument \u2014 a question.<\/p>\n Deciding which deals to abandon.<\/strong> A low score means “less like your past wins”, not “will not close”. Some of your best deals will be the ones that look unusual.<\/p>\n Replacing the pipeline review.<\/strong> A score does not know that the customer’s budget was frozen last week. Your salesperson does.<\/p>\n Forecasting revenue precisely.<\/strong> Aggregate scores give a rough direction. They do not give you a number to put in front of a board.<\/p>\n FIGURE 2: WHAT A SCORE IS AND IS NOT<\/p>\n A useful ordering<\/p>\n Not a prediction<\/p>\n This one is worth understanding, because it is subtle and self-reinforcing.<\/p>\n Salespeople stop working low-scored deals. Those deals then lose, because nobody worked them. The model sees they lost and becomes more confident that deals like that lose.<\/p>\n The score becomes true because people acted on it.<\/strong><\/p>\n Two safeguards:<\/p>\n Keep working a sample of low-scored deals.<\/strong> Not all of them \u2014 some proportion, deliberately. It keeps the data honest and occasionally finds business you would have missed.<\/p>\n Check calibration periodically.<\/strong> Of the deals scored highly, how many actually closed? Of the low-scored ones you did work, how many closed? If those numbers do not match the score’s implication, it is not learning your business properly.<\/p>\n Three habits, and they matter more than any setting.<\/p>\n Record losses with a reason.<\/strong> A short list \u2014 price, timing, competitor, no budget, no response. This is training data and it is also, separately, the most useful report in your CRM.<\/p>\n Keep stages accurate.<\/strong> Deals should move when something happens, not in a monthly tidy-up. A pipeline updated once a month teaches the model nothing about timing.<\/p>\n Close dead deals.<\/strong> An opportunity from eight months ago sitting in Negotiation is noise. Either work it or lose it.<\/p>\n FIGURE 3: WHAT DECIDES WHETHER THIS WORKS<\/p>\n Enough closed deals<\/p>\n Losses recorded properly<\/p>\n Stages that reflect reality<\/p>\n A sample of low scores still worked<\/p>\n Run it silently first.<\/strong> Score the pipeline for a quarter without showing the team. Then compare: did the high-scored deals actually close more often? You now know whether it works in your business rather than in general.<\/p>\n Show it as a sort order, not a number.<\/strong> A percentage next to a customer’s name invites arguments and false confidence. “Suggested priority” is a more honest presentation of what it is.<\/p>\n Keep the pipeline review.<\/strong> The score orders the conversation. It does not replace it.<\/p>\n Watch what it does to behaviour.<\/strong> If salespeople start ignoring whole categories of deal, that is the failure mode above starting.<\/p>\n Before switching it on, write down what should improve:<\/p>\n Conversion rate.<\/strong> Are more of the deals worked being won?<\/p>\n Time to close.<\/strong> Is attention going to deals that were ready?<\/p>\n Deals that go quiet.<\/strong> Does the stalled-deal alert actually catch things earlier?<\/p>\n Without a baseline you cannot answer any of these, and you will keep the feature on because it feels modern rather than because it works.<\/p>\n Sales prediction is a prioritisation tool<\/strong>. Used to order a salesperson’s day, it earns its place.<\/p>\n It needs real volume of closed deals, honestly recorded losses, and a pipeline that reflects what is actually happening. Without those three it produces a number that looks authoritative and means very little.<\/p>\n And keep working some of the low-scored deals<\/strong> \u2014 otherwise the score becomes right by making itself right.<\/p>\n Wondering whether your pipeline has enough history for this?<\/p>\n Get in touch. We will look at your closed-deal volume and how your losses are recorded, and tell you honestly whether scoring would help or just look impressive.<\/p>\n<\/div>","protected":false},"excerpt":{"rendered":" What it actually does Sales prediction does not tell you what will happen. It tells you which deals look most like the ones you have won before. That is a narrower claim than the marketing, and it is still useful. A salesperson with forty open opportunities and time for fifteen calls has a real prioritisation […]<\/p>\n","protected":false},"author":3,"featured_media":9175,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[26],"tags":[],"class_list":["post-9174","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai"],"yoast_head":"\nWhat it learns from<\/h2>\n
\n
\n
\n
\n
\n
The requirement nobody mentions<\/h2>\n
Where it genuinely helps<\/h2>\n
Where it does not<\/h2>\n
\n
\n
The failure mode to watch<\/h2>\n
Getting the data right<\/h2>\n
\n
\n
\n
\n
How to introduce it<\/h2>\n
Measuring it<\/h2>\n
The short version<\/h2>\n