# CRM analytics for an owner: five questions your CRM data answers before you buy a BI tool

https://salesbond.app/blog/crm-analytics-owner-five-questions-before-bi/
Published: 2026-10-28 · Updated: 2026-11-14

CRM analytics is the analysis of the deal, contact and activity data already in your CRM, and for a small team it starts with five questions: where the money sits by stage, what was won and lost and why, which deals are stuck, how the forecast compared with the result, and where deals come from. Each answer needs a handful of fields to be filled in. A BI tool comes later, once the answer lives in more than one system, needs history the CRM overwrites or needs calculations across groups of deals.

That order matters. Most teams that ask us about “CRM analytics” are not short of tools. They have reports nobody opens, because nobody wrote down the question the report was meant to answer.

## CRM analytics means two things: a method and a Salesforce product

Much of what Google returns for “CRM analytics” is about one product, Salesforce CRM Analytics, built natively on Salesforce. It is priced per user per month; some Sales, Service and Industry Cloud editions include it, and the others add it ( [Salesforce: CRM Analytics](https://www.salesforce.com/analytics/crm/)).

It is not Tableau, which Salesforce sells as a separate analytics platform. And if you already run Salesforce reports, you don’t need either one to start; the test for outgrowing built-in reports is the same in every CRM, and it is in the [last section](#when-built-in-crm-reports-stop-being-enough-and-you-need-a-bi-tool).

So if a consultant says “you need CRM Analytics” ask which one they mean: the license, or the habit of reading your own deal data. This article is about the habit, and it works with the reports Pipedrive, HubSpot, Salesforce and Zoho CRM already have.

## Question 1: where is the money, by stage and by owner

Open the pipeline, group it by stage, and sum the deal value. Then group the same deals by owner.

The first grouping shows where the pipeline is fat and where it is thin. Most of the value in the first stage is a list of hopes. Most of it in the last stage with nothing behind it is next month’s problem. The second grouping shows whether the number depends on one person. If one rep holds half the open value, that rep’s holiday is a business risk, and you want to know that before the holiday.

What makes this answer wrong is the value field. Deals created without a value sum to zero and silently disappear from the picture. Deals with a placeholder value, the round number a rep typed to get past a required field, inflate it. Before you trust the stage totals, filter for empty values and for suspiciously round ones, and read that list.

A sentence for the pipeline meeting: “We have X in stage three; which of those deals has a next step booked?”

## Question 2: what did we win and lose, and why

Pull every deal closed last quarter, won and lost, and look at the ratio and the reasons. The ratio by itself is dull. Won versus lost by stage is not. Most losses at the proposal stage point to price or qualification. Most losses at the first stage point to lead quality, and that is a different conversation with a different person. Win rate as a tracked metric is covered in [sales performance metrics and their CRM fields](https://salesbond.app/blog/sales-performance-metrics-crm-fields/).

The reasons depend on one field, and small teams usually find a hole here. In Pipedrive, lost reasons are freeform by default: any user types what they like, which gives you fifty variations of “no budget”. A user with the Deals admin permission can switch to a predefined list that reps choose from, and the account can hold up to 100 reasons ( [Pipedrive: lost reasons](https://support.pipedrive.com/en/article/lost-reasons)). We think the list should be short, under ten, with one “other” that requires a note. A long list is as useless as a free text box. Reps pick the first option that fits.

## Question 3: which deals are stuck and for how long

A stuck deal is one that has not moved stage and has no next activity scheduled. Both conditions. A deal with a demo booked for next week is only slow. A deal that a rep touched yesterday without booking anything is stuck, whatever the “last updated” date says.

That distinction is why “days in stage” on its own misleads. Reps update stuck deals, often because a manager asked about them, and the deal looks alive again. So don’t ask the data when the deal was last touched. Ask when the last activity with the customer was, and whether there is a next one. The activity log and the next activity field answer that; the updated timestamp does not.

Then sort by value. Five small stuck deals are a hygiene task. One stuck deal that is a third of the the quarter is a conversation today.

## Question 4: how did last quarter’s forecast compare with what closed

Owners ask this most and CRM data answers it worst, because a CRM stores the present and overwrites the past. When a rep moves the expected close date from March to April, the March forecast is gone. By the end of the quarter the CRM will tell you, honestly, that every open deal was always due next quarter.

So the answer needs the forecast as it stood at the start of the period, saved somewhere the CRM cannot overwrite: a spreadsheet with a column per month, or a snapshot feature where the CRM has one (here is how [reporting snapshots in Salesforce](https://salesbond.app/blog/salesforce-reporting-snapshots-pipeline-trend/) work).

With the saved forecast in hand, the comparison is three columns: what we said, what closed, and what happened to the difference (pushed, lost, or still open). We think “pushed” is the most useful number here, because it shows how far the team’s close dates can be trusted. If half the forecast moved to next quarter, the close date field is a wish, and every forecast built on it inherits that.

## Question 5: where do deals come from

Group won deals by source and compare value. Referrals may be a small share of leads and a large share of revenue. Sum the value next to the count, because the source that keeps the team busiest and the source that brings the most revenue are often different.

The field here is usually called source, lead source or channel, and it has the same disease as lost reason. If the list is open, reps fill it inconsistently, and if it is long, they stop filling it. Keep it to the channels you spend money or time on, and make it required at creation. Nobody remembers in June where a January lead came from.

Source records the first touch, and for a small team thats enough. Multi-touch attribution is one of the signs the question has outgrown the CRM.

## Fields every question depends on, and how bad data breaks the answer

Five questions, seven fields: deal value, stage, owner, lost reason, the activity fields (last and next), expected close date, and source. Not one of them is exotic.

![Diagram linking five CRM questions to the seven fields they use, with how each field's data goes wrong](https://salesbond.app/wp-content/uploads/2026/10/crm-analytics-owner-five-questions-before-bi-2.png)

Answers usually break on wrong data in the few deals that carry the most weight, and an empty value on the biggest deal moves the stage total more than a hundred empty values on small ones.

So skip the general clean-up and run one filter per question: no value, no lost reason, no next activity, a close date that moved, no source. Each filter is a list of deals with a name next to it. Read the list together with the rep. The [metrics that make CRM data quality visible](https://salesbond.app/blog/crm-data-quality-metrics/) go deeper, and the weekly [CRM reports](https://salesbond.app/blog/crm-reporting-owner-head-of-sales-report-set-weekly-cadence/) and [dashboard tiles](https://salesbond.app/blog/crm-dashboard-owner-head-of-sales-tiles-and-fields/) for an owner put the same questions on a calendar. SalesBond’s free CRM audit runs a read like this on the last 30 days of a connected CRM and sends a PDF report with money at risk, deal health, the pipeline bottleneck and what to fix, then a second PDF seven days later with what changed.

## When built-in CRM reports stop being enough and you need a BI tool

You need a BI tool when the answer lives in more than one system, when you need history the CRM overwrites, or when you need calculations across groups of deals; company size is not the trigger. A BI tool here means software that pulls data from several sources into one place, keeps it over time and lets you build your own calculations on top, and [business intelligence tools for a sales team](https://salesbond.app/blog/business-intelligence-tools-for-a-sales-team-compared-by-job-no-ranking/) are compared by the job each one does, with no ranking, in a separate guide.

Several systems first: revenue in the accounting tool, deals in the CRM, usage in the product database. The moment “which customers paid late and also have an open renewal” is a real question, you are joining two systems, and that is BI work.

Then history. If you keep rebuilding “how did this look three months ago” by hand from exports, you need stored snapshots, and a CRM report builder does not keep them.

Then arithmetic: weighted pipeline by your own probabilities, rep capacity, win rates by quarter of creation. Report builders group and sum but are poor at calculations between groups, and a spreadsheet gets fragile once a second person edits it.

If none of the three applies, don’t buy the tool yet. Fix the fields first, because a BI dashboard built on empty lost reasons charts the same gaps at a higher price. The same reasoning runs through [Zoho Analytics versus Zoho CRM reports](https://salesbond.app/blog/zoho-analytics-vs-zoho-crm-reports/) and [revenue intelligence from CRM data first](https://salesbond.app/blog/revenue-intelligence-small-sales-team-crm-data-first/), and each of the four types in [business analytics for a small sales team](https://salesbond.app/blog/business-analytics-small-sales-team-four-types-with-crm-data/) starts from CRM fields you already have.

## FAQ

### Is CRM Analytics the same as Tableau?

No. Both are Salesforce products, but CRM Analytics is built natively on Salesforce and works inside the CRM, while Tableau is a separate analytics platform (Tableau Cloud, Tableau Server, Tableau Next) ( [Salesforce](https://www.salesforce.com/analytics/crm/)).

### What does a CRM analyst do?

A CRM analyst turns CRM data into answers sales and management can act on. In a company with five reps, that is usually the owner, on Monday morning, for twenty minutes, with the five questions above.

### Do I need a BI tool if my CRM has reports?

Usually not yet. A BI tool starts to pay off when, for example, you want last quarter’s forecast next to what closed and the CRM has already overwritten the old close dates. Until then, fill in the seven fields above.
