Getting Started with Count

:wave:This guide will show you how to get started making exciting discoveries using Count. Don’t forget you can ask questions in our help centre too!

Exploring the data

When you open Count, the data is already loaded. Click on View Dataset to see the available tables and how they all connect.

The tables and their columns are listed on the left-hand side as a a reminder of the data you can explore.
You can preview each data table by clicking View Table to open the relevant data table.

Shared visuals

In the Shared visuals section, discover what others have already found.

Open up the visual to see it in more detail and click Add to Notebook to add the insight to the Explore section.

The insight will be added to next the available cell where you can then edit and update the query.
(A cell looks like a search bar and is where you will type your requests. You can add as many queries as you like, building up your notebook down the page.)

Start making your own discoveries in the Notebook section

There are just four simple rules for using Count.

Rule 1: Select data by naming the column(s) you want

  • Whenever you want to select data, simply type the table name and column name of the columns you want in a cell using the format TableName.ColumnName e.g. Company.company_id,, Industry.industry
  • You can request as many columns as you like, regardless of whether or not they are from the same table. Count will detect which table the column comes from and connect tables automatically.
  • Count helps you by autocompleting what you are typing and correcting for spelling errors.

  • Check that the results are what you expected by looking at Showing Results and the column headers of the results table.

  • You can also select an entire table by typing TableName.*

  • Open the results table and sort it in ascending or descending order by clicking on the header of the relevant column.

Rule 2: Functions come before the relevant column name

  • To do a calculation, you need to type the function name you want to use before the applicable column. e.g. Industry.industry, sum(total_funding), count( desc, avg(Company.valuation), year(Company.valuation_date)

  • Common functions to use on number columns include sum, count, distinct, max, min, average

  • Common functions to use on text columns include count, length

  • You can use functions to with date/time columns to specify the unit of time you want from a relevant column, e.g. month, year, day, month_of_year, week_of_year, day_of_week

  • A full list of functions can be found in the Common functions help section by clicking on the help button in the bottom right of the page. Or, you can click here.

Rule 3: Filter results by specifying what you want to filter for after the relevant column name.

  • You can filter data using logic expressions. e.g. Company.hq_city = 'London’

  • Comparisons can be performed using =, <, >, !=, ≥, ≤ to specify how a given column should be filtered e.g. Company.employees > 30

  • You can filter for fragments of text in a text column by using CONTAINS(ColumnName,‘example fragment’) e.g. COUNT(Company.company_id), CONTAINS(Company.revenue_model,‘advertising’)

  • NB: if you want to include the filtered value in the results table, you will need to repeat the column name, e.g. COUNT(Company.company_id), Company.revenue_model, CONTAINS(Company.revenue_model,‘advertising’)

  • The output table can also be filtered by typing distinct next to the relevant column name within the query to remove any duplicate rows

  • The Limit button reduces the number of results returned. By default it returns only the first 10 rows and the number of results returned can be specified by replacing 10 in the query cell with the required number.

  • By default all filters are inclusive (like AND logic if you are familiar with this) meaning a row has to pass all the filter conditions to be included. You can use the word OR (in capitals) between different conditions if you want rows to be included if they meet one or more of any of the conditions.

Rule 4: Refer to previous results using the cell’s letter reference as the table name.

  • You can refer to previous results (tables or columns) using the cell’s letter reference as the table name (e.g. m, n, o)
  • You can either use the tableletter.columnname or the tableletter.columnnumber, e.g. select the third column of cell a using a.column_name, or by using the column number as shorthand a.3
  • All the following cells which refer back to a particular cell will update automatically if you change the cell they are referring to.

Share your insights

Share your findings to the insights section or on social media by clicking the share button at the top of your visual. Add a short description to explain what your chart shows.