Sales dashboard: build one your team actually acts on

A sales dashboard is a single view of the numbers your team uses to decide what to do next week. Not a report, not an archive: a decision surface. The useful ones hold fewer than ten metrics, each tied to an action someone can actually take. The useless ones hold forty, get opened once a quarter, and quietly become the reason nobody trusts the data any more.

This article covers which metrics earn a slot, which ones mislead, how to split the view by who is reading it, and the layer most dashboards miss entirely.

What a sales dashboard is for

A sales dashboard answers one question: is the pipeline going to produce the number, and if not, where is it leaking?

That framing rules a lot out. A metric earns its place when the person looking at it can do something differently on Monday morning. If the answer to "and then what?" is silence, the metric is interesting rather than useful, and interesting belongs elsewhere.

Three things a dashboard is not:

• A reporting archive. History and variance analysis live in reports. A dashboard shows the present and the immediate future.
• A performance review. Evaluating one person against weighted criteria is a sales scorecard, a different object with a different cadence and a different conversation attached to it.
• A forecast. A forecast is an output. The dashboard shows the inputs that make it credible, or expose it as wishful thinking.

One more distinction worth holding. A dashboard answers what the numbers are, week after week. If your question is why they are down, that is a sales performance diagnosis, a one-off exercise with a different method and a different set of numbers.

The metrics that earn a slot on a sales dashboard

Nine metrics cover almost every B2B sales team. Most teams need fewer. The test in the third column is the one that matters: if you cannot fill it in, cut the metric.

The last two are hygiene metrics, and they are the ones teams skip. They deserve a slot because they tell you whether the other seven mean anything. A pipeline where a third of open deals have no next step and no budget confirmed is not a pipeline you can read.

If you want a wider catalogue to pick from, HubSpot keeps a long list of sales dashboard examples and Salesforce publishes its own set of nine sales KPIs. Both are worth skimming, and both will tempt you into putting too much on one screen.

Three metrics that mislead more than they help

Some numbers look like signal and behave like noise.

Activity volume on its own. Calls made, emails sent, meetings booked. These measure effort, not progress, and they are the easiest numbers in the stack to game. A rep who makes 80 calls into the wrong segment outranks a rep who makes 30 into the right one. Activity is useful as a denominator (connect rate, meetings per 100 calls) and misleading as a headline.

Pipeline value with no age or stage weighting. A two-million pipeline in which half the deals have not moved in two months is not a two-million pipeline. Either weight by stage probability, or show pipeline alongside the stuck-deal count. On its own, the total is the single most comforting and least informative number on most dashboards.

Individual win rate on small samples. A rep who closed three of five deals is at 60%, and that figure means nothing. Below roughly twenty closed opportunities, individual win rate moves more with luck than with skill. Aggregate it at team level, or look at stage conversion instead.

Sales KPIs, sales metrics and a scorecard are not the same object

These terms get used interchangeably, and the confusion is why dashboards bloat.

• A sales metric is any number you can measure. There are hundreds.
• A sales KPI is the small subset you have decided to steer by. A KPI is a metric with a target, an owner and a review date. Without those three, it is just a metric.
• A sales dashboard is the view that displays the KPIs, in the timeframe the reader works in.
• A sales scorecard evaluates one person against criteria you weight on purpose. It feeds coaching, not pipeline decisions.

The practical consequence: a dashboard with twenty metrics has no KPIs, because nothing on it has been prioritised. Picking what to leave out is the whole exercise.

One dashboard per audience, not one for everyone

The single most common mistake is building one dashboard and showing it to three audiences who work in three different timeframes.

Reps ignore dashboards built for leadership, and leadership drowns in dashboards built for reps. Three focused views beat one comprehensive one, and they are cheaper to maintain than they look: the underlying metrics are mostly the same, filtered and aggregated differently.

The cadence that turns a dashboard into a decision

A dashboard with no meeting attached to it gets checked for a fortnight and then forgotten. Tie each view to something already on the calendar.

1. Weekly pipeline review, rep view. Fifteen minutes. One question per deal with no next step, one per deal stuck more than fourteen days. Nothing else.
2. Monthly, manager view. Stage conversion and cycle length. The question is where the process is leaking, not who is behind.
3. Quarterly, leadership view. Trends only. A quarter is the shortest window in which a trend in win rate or cycle length means anything.

One rule makes the whole thing stick: every metric on the dashboard has exactly one named owner. A number nobody owns is a number nobody fixes, and after two quarters of nobody fixing it, people stop looking at the dashboard altogether.

Why most sales dashboards only show half the deal

Everything above comes out of the CRM. And the CRM only knows what someone typed into it.

So the dashboard shows what happened to a deal (stage, amount, close date, owner) and almost nothing about why. The budget objection raised on the second call, the competitor named in passing, the fact that the economic buyer has still not joined a single meeting: none of that is a field anyone fills in reliably, so none of it reaches the screen.

Two consequences follow, and they are the reason dashboards lose credibility.

First, the metrics that depend on conversation content are the first to break. Qualification completeness, next step quality, risk signals: all of them rest on a rep remembering to log something after a call, which happens on good weeks and not on busy ones. The hygiene metrics that tell you whether the dashboard is trustworthy are themselves the least trustworthy numbers on it.

Second, the dashboard inherits every gap in the CRM. No chart fixes a missing field. The problem is upstream, in how the data gets captured in the first place, which is a data quality problem rather than a reporting problem.

What closes the gap is a layer that turns the conversations themselves into structured fields, so the dashboard reads data that was captured rather than data that was remembered. That is the job call analysis does, and it is worth understanding before you compare sales analytics tools: most of them visualise your CRM beautifully without changing what goes into it.

Where Praiz fits in

Praiz sits between the conversations and the CRM as an infrastructure layer. Specialised AI agents work on every call and meeting (video, phone, in person) and write structured data back: the qualification criteria, the objections raised, the competitors named, the next step agreed, the risk signals. Three families of agents cover it: generation, scoring and tracking.

For a dashboard, that changes where the numbers come from. The fields behind the hygiene metrics get filled without anyone typing them. Teams running Praiz see their CRM completion rate multiply by five, with 90% reliability on the strategic CRM fields, and 100% of deals carrying a completed MEDDIC against 32% before. Every call is scored rather than a sample of them, so quality metrics stop depending on who had time to review.

The agents are configurable rather than fixed, which matters because your qualification criteria are not generic: you can start from the Praiz AI agents library or build your own, and map each field to your CRM. The Praiz team sets this up with you instead of handing you a blank configuration screen. That is worth asking any vendor about: what onboarding they provide, who does the configuration, and how fast support answers. A dashboard fed by a tool nobody set up properly shows exactly the same gaps as the one you have now.

How to build your first sales dashboard in a week

Start narrower than feels comfortable. Dashboards grow on their own; they never shrink on their own.

1. Write down the one decision it has to support. One sentence. "Decide which deals to push and which to drop in the weekly review" is a decision. "Give visibility on performance" is not.
2. Pick five metrics, not nine. Coverage, qualified opportunities created, stage conversion, deals with no next step, stuck deals. You can add later; you will not remove later.
3. Run each one through "and then what?" If nobody can name the action, cut it. Do this before you build anything, while cutting is free.
4. Attach it to an existing meeting. The weekly pipeline review, as the first agenda item. A dashboard introduced without a meeting has roughly two weeks to live.
5. Name an owner per metric, and delete one metric a month until someone complains. The complaint tells you where the floor is. Until then, you are carrying weight.

Salesforce has a useful set of dashboard layouts by role if you want a starting template, though the same caution applies: a template shows you what is possible, not what you need.

Frequently asked questions