Sales pipeline

The image is literal: deals enter one end of a pipe, move from stage to stage and leave the other end, won or lost. The sales pipeline is the list of those deals with the current stage of each one.

It is usually drawn as columns, like a Kanban board: one card per deal, one column per stage. A glance tells you whether the meetings column is emptying or proposals are piling up without answers.

A pipeline is not the same as a sales funnel. The funnel describes the typical journey and its conversion rates. The pipeline shows actual deals, today, with names, amounts and dates.

There is no official list. A common B2B template: prospect identified, first contact, conversation in progress, meeting scheduled, need qualified, proposal sent, negotiation, won or lost. An agency selling short projects will have fewer stages, a software vendor selling to large enterprises will have more.

The rule that matters: every stage must match a verifiable fact, not a feeling. 'Meeting scheduled' can be checked in your calendar. 'Very hot prospect' cannot be checked anywhere. If two reps could put the same deal in two different columns, your definition is too vague.

Add a last activity date to every card. A deal that has not moved in thirty days is not really in the pipeline anymore.

Four measures are enough to steer: the number of deals per stage, the total pipeline value, the conversion rate from one stage to the next and the average time spent in each stage. With those numbers you know where to act.

Pipeline velocity combines them. The common formula: number of opportunities times average deal value times win rate, divided by the sales cycle length in days. The result is revenue per day. Use it to compare two quarters or two prospecting methods, not to forecast to the cent.

For forecasting, weight each deal by a probability tied to its stage, for example 10% at the first meeting and 60% in negotiation, then correct those percentages with your real history.

Finally, watch coverage: pipeline value compared with your revenue target for the same period. If your quarterly target is 60,000 and you close about one deal in four, you need at least 240,000 of qualified deals in progress. Below that, the problem sits at the top of the pipeline, and the time to prospect is now, not at the end of the quarter.

The most widespread one is never closing lost deals. The pipeline swells, the forecast with it, and the team reassures itself with false numbers. Mark a deal lost as soon as the prospect says no or has gone quiet for a long time, and record the reason.

Another is watching only the bottom of the pipeline. If you only look at open proposals, you find out too late that first contacts dried up two months ago. The top of the pipeline fills slowly and empties without a sound.

Last, many pipelines depend on manual updates. A busy rep does not update the CRM, and the view goes stale within weeks.

LinkedIn mostly feeds the top of the pipeline: first contact, conversation, meeting. Work the math backward. If you want four meetings a month and it takes about 60 invitations to strangers for one meeting (1.6% observed meetings in MimikFlow's 2026 LinkedIn Prospecting Observatory, published in French), you need around 250 invitations a month, roughly 60 a week, before counting warmer sources.

MimikFlow keeps that part of the pipeline current with no data entry. Every prospect carries a status that follows what really happened: invitation sent, accepted, first message, conversation, goal reached. The Pipeline view of the Prospects page shows those automatic columns, and you can create your own sales columns, such as 'proposal' or 'negotiation', then drag cards into them. Moving a card does not change the automation, and the filtered list can be exported to your CRM.

LinkedIn prospecting has its own rhythm. Invitations play out fast, but replies come slowly: in the 2026 Observatory, 15.2% of replies arrived after the first week. A pipeline fed by LinkedIn needs a steady weekly input so meetings keep landing at a regular pace.