Sales Velocity

Sales velocity expresses a pipeline as a rate rather than a total: how much revenue it produces per day. That reframing is the point. A pipeline worth 4 million dollars tells you little on its own, because the same total behaves differently depending on how many deals it contains, how often they close, and how long they take.

The formula

Multiply the number of qualified opportunities by the average deal value, multiply that by the win rate, then divide by the average sales cycle length in days. A team carrying 60 qualified opportunities at an average of 30,000 dollars, winning 25 percent, on a 90 day cycle, has a sales velocity of 5,000 dollars per day. The output is not a forecast. It is a diagnostic that shows which of the four inputs is constraining revenue.

Sales velocity and pipeline velocity

The two terms are generally used interchangeably for this calculation. Where teams do draw a distinction, pipeline velocity is sometimes used more narrowly for the speed at which deals move between stages, while sales velocity keeps the full four-input formula. Because the usage is not consistent across sources, it is worth stating which definition a dashboard uses rather than assuming the reader shares it.

Reading the four levers

Each input responds to different work. Opportunity count is a demand and qualification problem, addressed upstream at lead conversion. Average deal value responds to packaging, segment focus and multi-product attach. Win rate responds to qualification discipline and competitive positioning. Cycle length responds to process friction: approvals, security review, procurement and quoting.

The lever most teams reach for first is opportunity count, because more pipeline feels like the obvious answer. It is often the least efficient of the four. Halving a 90 day cycle doubles velocity without adding a single deal, and cycle length is frequently the input with the most internal, controllable friction in it.

Measuring it honestly in Salesforce

All four inputs come from the Opportunity object, which makes the calculation straightforward and the data quality question hard. Cycle length depends on CreatedDate and CloseDate being real rather than backfilled. Win rate depends on lost deals being marked lost rather than left open. Average deal value distorts quickly when one outlier deal sits inside a small sample, so a median is often more honest than a mean.

The deeper problem is that the fields describing why a deal moved slowly are usually empty. Salesforce State of Sales 2024 found reps spend 72 percent of their time not selling, and a large share of that is administrative capture that gets skipped under pressure. When 60 percent of sales activity never gets logged, cycle length is measurable but unexplainable. GPTfy addresses that by drafting the record from the call transcripts, emails and meeting notes already attached to the Opportunity, so the diagnosis has evidence behind it.