How to Analyze a HubSpot Deals Export in Excel

HubSpot's built-in reports are fine for the dashboard view: deals by stage, a forecast, a leaderboard. Once you have a sharper question, like which rep closes fastest, which source brings deals that actually win, or which open deals have gone quiet, you usually end up exporting the deals to a spreadsheet. The export is tidy, but a few conventions (open vs closed deals, blank amounts, several pipelines, currencies) will skew your numbers if you don't deal with them first. This guide covers how to get a useful export, the columns that matter, the traps, and the questions worth asking.

Getting a deals export that has the right columns

In HubSpot, go to CRM → Deals and switch to the table (list) view. The export uses the view you're on: the filters decide which deals are included, and the columns decide which properties you get. So set it up first:

1. Remove any filters that hide closed deals, unless you only want the open pipeline. Win-rate analysis needs closed-won and closed-lost deals.
2. Use Edit columns to add the properties you'll need (see below).
3. Click Export, choose CSV, XLS, or XLSX, and decide whether to include only the columns in the view or all properties. HubSpot emails you a download link when the file is ready.

"All properties" makes a wide file, but it saves you a second export when a question needs a column you forgot. Most people pick all properties once, then trim.

If you also want to analyze by lead source or company size, export contacts or companies the same way. Record IDs link them: each deal row can include associated contact or company IDs, and a spreadsheet with a deals sheet and a contacts sheet lets you join on those.

The columns that matter

The exact labels depend on your portal and any custom properties, but a standard deals export has these:

Depending on your setup, you may also see Is Closed Won, Is Deal Closed?, Deal probability, Forecast amount, Currency, Amount in company currency, Last Activity Date, and source properties such as Original Traffic Source. Add them if they're there; they make the analysis much easier.

Traps that change your numbers

Open deals are not losses. A win rate of won ÷ all deals punishes every deal that's still in progress. The honest definition is closed-won ÷ (closed-won + closed-lost), counting only deals that have actually closed. If your export has Is Deal Closed? and Is Closed Won, use them; otherwise map each Deal Stage label to open, won, or lost yourself.

Close Date means two things. On a closed deal it's when the deal closed. On an open deal it's the rep's estimate. Group closed deals by Close Date for "what did we win in March," but label open deals by Close Date as a forecast. And watch for open deals whose close date is already in the past. That's a stalled-deal signal, not a data error.

Blank amounts. Many teams only fill in Amount once a deal is qualified. Summing Amount silently treats those deals as zero; averaging it silently drops them. Count how many deals have no amount before you report pipeline value, and state it next to the total.

Multiple pipelines. If you run separate pipelines (new business, renewals, partner deals), their stages and cycle lengths aren't comparable. A "Proposal" stage can exist in two pipelines with different meanings. Filter or group by Pipeline before anything else.

Currency. In a multi-currency portal, Amount is in the deal's own currency. Summing it across currencies gives a meaningless number. Use Amount in company currency if your export has it, or group by Currency.

Stage history isn't in the default export. Deal Stage is the current stage only. The dates each deal entered each stage aren't part of a normal deals export, so time-in-stage analysis needs either the stage date properties your portal exposes or HubSpot's property history export. The standard file still supports overall cycle length (Create Date to Close Date).

Dates and formats. Dates export as text in the format your account uses. A spreadsheet that reads 03/04/2026 as April 3 when you meant March 4 will scramble your monthly numbers. Check a few rows against HubSpot before trusting any trend.

Questions worth asking

Once the file is clean, these are the questions that tend to change what a sales team does:

1. Win rate by owner. Closed-won ÷ closed deals per Deal Owner, with the deal count next to it. A 60% rate on five deals isn't the same as 40% on eighty.
2. Win rate by source. Group by the original source (from the deal, or joined from the contact export). Some channels bring volume, others bring deals that close.
3. Sales cycle length. Median days from Create Date to Close Date for closed-won deals, by owner or by deal size band. Median rather than mean, because one deal that sat for a year distorts the average.
4. Pipeline by stage and month. Open deals only, sum of Amount by Deal Stage and by expected Close Date month. Split by Pipeline if you have several.
5. Stalled deals. Open deals whose Close Date has passed, or whose Last Activity Date is more than 30 days old. Sorted by Amount, that's the list for your next pipeline review.
6. Average deal size, won vs lost. If lost deals are consistently bigger, you're either reaching above your fit or losing on price.
7. Loss reasons. If your team fills in Closed Lost Reason, count the reasons by owner and by stage. It's often the most useful free-text field in the export.

Doing it without formulas

Every question above comes down to filtering and grouping: the kind of work you'd otherwise build a pivot table for. With ChatExcel you upload the export (Excel or CSV, and every sheet is read, so a deals sheet and a contacts sheet can go in one workbook) and ask in plain English:

• "Win rate by deal owner, counting only closed-won and closed-lost deals, with the number of closed deals."
• "Median days from create date to close date for won deals, by owner."
• "Open pipeline by deal stage and expected close month, as a bar chart."
• "List open deals with a close date in the past, largest amount first."

Answers are calculated from every row in the file, not a sample, and you can turn any result into an interactive bar, line, area, or pie chart. Put the ground rules in your first question ("treat closed-lost as a loss, ignore open deals, only the New Business pipeline") so they carry through. When you want to rebuild a number in your own workbook, ask "what's the Excel formula for that?" and you'll get the COUNTIFS or AVERAGEIFS version. If the export needs tidying first, the data cleaning guide covers blank rows, stray headers, and mixed date formats.

A quick sanity check

Before you share a number, compare two things with HubSpot itself: the count of closed-won deals in a month, and total closed-won Amount for the same month. If they match the HubSpot deal report for that period, your filters are right. If they don't, it's almost always a pipeline you forgot to filter, a currency mix, or deals whose Close Date was edited after they closed.