Two companies can share a name. An old website can lead somewhere unexpected. A search result can look convincing while describing an organization you never meant to research. Finding people at the wrong company is a lot of work for nothing, however tidy the resulting contact list looks.

So Contact Finder starts with the organization. Each match brings together the website, the company LinkedIn page, the information supporting the match and the reason the company belongs in the Maximo research, all laid out for review.

Companies are presented in groups of ten. A reviewer can approve a match, remove an entry or correct the company, website or LinkedIn details and run it again. Ten is a manageable number to inspect properly, and only approved companies move on to contact research.

Findings stay attached to the search that produced them, so a later session picks up from the decisions already made.

Contact details need more care than a neat row suggests. A personal LinkedIn profile has to point to someone the research supports; a company page serves a different purpose.

Email addresses come in distinct kinds. An address the person published is different from a general company inbox, and both differ from one inferred from the company's naming pattern. A likely first.last address can stay on the list as a lead, labeled as an assumption.

The application keeps runs, corrections and the material behind each finding, and exports to CSV when the research needs to go elsewhere. It also separates a completed search that found no contacts from a company nobody has processed yet. My work covered the research process, the interface, the Banetti presentation and Google Workspace sign-in.