Lesson 1 of 7

Lesson 2.1.1

Website Finder first, always, and why everything depends on it

5 min readIntermediate

What this costs

A flat charge for every row it processes, so run it only on the rows that are missing a website. The figure is on the tool card.

Before you start

A table with company names.

The official company website is the foundational data point in Kuration. It is what people tools, technology lookups and most company research resolve against. A row without one is a row that will silently skip half your enrichment stack, and you will not find out until you wonder why the coverage is poor.

The dependency almost everything has

Free filters0 creditsWebsitethe dependencyScore0 creditsPeoplethe expensive oneEach step runs only on the rows the step before it kept1,000 rows in, about 150 rows reach the expensive columnEach column waits for the column it reads, through its dependencies

Company name is ambiguous. Three companies share a name, one has changed it, one trades under something else entirely. A domain is unique. Every tool that has to be sure it is looking at the right company wants the domain, which is why the order is always: names, then domains, then everything else.

  1. 1

    Check what proportion of rows already have a website

    Google Maps gives you one for many rows. A directory extraction often does too. Those rows never need Website Finder, so the job is to run it on the gaps only.

  2. 2

    Fill the gaps with Enrich Empty Cells

    If the websites you have already sit in a column, open that column's menu with the three dots in its header and choose Enrich Empty Cells, then pick Official Website Finder. The column becomes a tool column, and when you run it only the empty cells are filled. The websites you already had are skipped, so you never pay to find one you hold.

  3. 3

    No website column yet? Add Website Finder and run it on the rows you want

    Open the Toolbox, choose Official Website Finder and map the company name column. Every cell in a new column starts empty, so filter to the rows you want, or set a condition, before you run it.

  4. 4

    Pass anything that tells two companies apart

    Location, industry, company email and company phone are all optional inputs, and each one raises accuracy when company names are common. The email's domain and a phone number that matches a candidate page both count as evidence. If your source gave you a city, pass it.

  5. 5

    Filter out the rows that still have no website

    Some companies genuinely do not have one. Decide now whether they are worth a different treatment or worth cutting, but do not carry them into a people search.

Paid to find websites you already had

What you see
Website Finder added as a new column and run on every row
The fix
Use Enrich Empty Cells on the website column you already have, so only the blank cells run. Or filter to the rows with no website before running a new column.

Common company name, wrong domain returned

What you see
A domain that clearly belongs to a different company
The fix
Pass location, industry, company email or company phone as optional inputs. They exist for exactly this.

Carried website-less rows forward

What you see
Poor coverage on every downstream tool
The fix
Filter them out or route them separately before enriching.

Check yourself

0 of 2 answered

1.Why does the official website come before everything else?

2.Your website column is already filled on 480 of 1,000 rows. What is the cheapest way to fill the other 520?

Answer all 2, then open Conditions: run a column only on rows that qualify to tick this lesson off.

Try it yourself

On a list of at least 100 rows where some websites are already filled, use Enrich Empty Cells on the website column with Official Website Finder and run it. Compare the number of cells it ran on with the size of the table.

You have done it when
Only the empty cells ran, and you can state the saving against a run on every row as a number.
Credit budget
The card price, on the empty cells only.