Why a data quality dashboard for Cirqll?

Anyone who builds a dashboard on Cirqll usually discovers in the first week what the data really looks like. A customer that appears three times counts three times in your revenue per customer. A deal without a value makes your pipeline look smaller than it is. And an automation that recognises a customer by email address will happily create a new one when the email field is empty.

That is nobody's fault. A CRM is filled by several people, in busy weeks and with different habits. Mess builds up by itself. A one-off clean-up helps for a while, but without visibility of the quality it drifts again within a few months.

A data quality dashboard turns that around. Every week it shows what is wrong, who owns it and whether it is getting better or worse. Cleaning up becomes a small, regular task instead of a yearly project.

Which data quality checks do you run on Cirqll?

You do not need to measure everything. Look at the data your reports and automations rely on. For most Cirqll users these are the five checks:

  1. Duplicate customers. Customers with the same Chamber of Commerce (KvK) number, the same email address or an almost identical name. This is the most important check, because duplicates pollute everything that follows.
  2. Missing core data. Customers without a KvK number, email address or phone number, and customers without a contact person. Without these you cannot reliably recognise a customer or match it to your accounting.
  3. Incomplete deals. Open deals without a value, without a percentage or assigned to a user who is no longer active. They make your pipeline and forecast unreliable.
  4. Stalled customers and deals. Customers with no activity for a long time and no next contact planned, and open deals that have not changed in months. These are often deals that are effectively lost but were never closed.
  5. Empty custom fields. Fields you created yourself, such as lead source or sector, that are only half filled in. Those are exactly the fields that give your reports meaning.

Each of these checks is a simple count: how many records fail, and what percentage of the total is that.

Which Cirqll data do you need for it?

Everything you need comes from the Cirqll API. When we checked through the API on 27 October 2026, we saw the following data, among other things:

  • For customers: name, KvK number, email address, phone number, owner, priority, a next contact date and the date of the last activity. You can also request the number of contacts and deals per customer.
  • For deals: name, value, percentage, customer, owner, the created and last-changed dates, and whether the deal has been archived.
  • Custom fields: for each field you can see which part it belongs to (for example customer or contact), what type it is (such as yes/no, dropdown, labels, date or amount) and whether it is required.

That last point is useful: a required field that is still empty often points to records that came in through an import or an integration. What else you can do with the API is covered in Cirqll API: what you can do with it.

How do you build a Cirqll data quality dashboard in Power BI?

If you already have a Power BI report on Cirqll, you build this dashboard on the same connection. How that connection works is explained in connect Cirqll to Power BI. A layout that works well in practice:

  1. One page with the current position. A card per check with the count and the percentage: so many duplicate customers, so many customers without an email address, so many open deals without a value. Red when it exceeds an agreed threshold.
  2. One page per owner. The same checks, split by user. Not to blame anyone, but so everyone can see what is on their own plate.
  3. A work list. A table of the specific records that need attention, with a link or ID so you can find them in Cirqll straight away. This is the page people actually use.
  4. A trend. How does the percentage develop week by week? For that you need to store the results of the checks. You can do that in a database of your own, as described in Cirqll data in your own database.

For duplicate customers a few rules are enough to start with. Compare KvK numbers and email addresses exactly, and compare names after removing capitals, spaces and suffixes such as "B.V.". That way you find "Bakkerij Jansen B.V." and "bakkerij jansen" as a possible duplicate. Show this as "possible duplicate": the decision to merge is made by a person.

An example: cleaning up before you automate

A fictional but recognisable example. An installation company with three salespeople wants web requests to go into Cirqll automatically. The automation recognises an existing customer by KvK number or email address. Before it goes live, they first build the data quality dashboard.

The first picture is sobering. A sizeable share of customers has no KvK number, a number of customers appear twice, and there are open deals that have not been touched in more than six months. In this situation the automation would keep creating duplicate customers.

The approach:

  1. First the possible duplicates from the work list. Each salesperson goes through their own list and merges or deletes.
  2. Then the old deals: for each one, the question whether it is still alive. If not, close it. The pipeline gets smaller, but honest.
  3. Then the missing KvK numbers, starting with customers that have open deals.
  4. Only when the percentages are below the agreed threshold does the automation go live.

The dashboard keeps running after that. Every Monday the team spends five minutes on it. How to decide which automation comes next is covered in automating Cirqll: which manual tasks to remove first.

How do you keep your Cirqll data clean afterwards?

A dashboard shows what is wrong. Preventing it is even better. A few measures that help:

  • Make the key fields required. In Cirqll you can make custom fields required. Only do this for fields you actually use; ten required fields mainly lead to fake input.
  • Agree how you recognise a customer. For example: KvK number first, then email address. Build that rule into your automations as well, so they do not create duplicates themselves.
  • Let automations do the boring work. With Make.com you can, for example, send each owner a weekly email with their own work list, or create a task when a new customer is added without a KvK number. Examples of such scenarios are in Cirqll and Make.com: 5 automations.
  • Look at it regularly. Make data quality a fixed item on your sales meeting agenda. Five minutes a week is enough.

Help with your Cirqll data quality dashboard

We usually build a data quality dashboard as part of a Cirqll report in Power BI. See what we do in dashboarding, read more about our partnership with Cirqll and the Cirqll Power BI Connector on our Cirqll page, or see how workflow automation keeps your data clean. Which other systems we connect is shown under integrations.