The Search from within - we aren't there yet

Searching and finding documents and data has been an ongoing struggle for many companies and government agencies.

PDF files on a shared drive were how many entities managed their documents. Then came intelligent capture and document management.

Some of the reasons why companies/agencies still have trouble finding what they need are:

• Bad file naming conventions

• Lack of good metadata and indexing practices

• Inadequate user training

• Proprietary databases

• Files being saved to the wrong places

• Too many applications that are not talking to each other

If it was possible to run a search through every application, database, and server, it would be a game-changer. It would also be a data privacy and governance nightmare.

Why?

Searching through documents and data is not the same as doing a Google search or ChatGPT prompt. These search tools are for searching the internet and there are currently no controls on what they bring back to the searcher/prompter. These may be great tools for marketing departments and research, but not useful to departments like Accounting, HR, Planning, Law Enforcement, etc.

The reason there are so many applications used with separate databases, is that departments and users need different tools to do their jobs. And, controlling who has access to the applications and the data and documents within them is crucial.

There is also the question of who has the right to add new documents or data to an application or server. There are plenty of companies advertising AI office automation tools, but how they get those new documents/data into a third-party application is another story.

It would be great to have one search tool that we could use that would talk to all the applications we need it to, but that seems unlikely. Most of us have experienced scenarios where we see duplicate data going into multiple applications and wonder why they are not integrated. And questioning why we have to go here to search for this and there to search for something else.

There definitely is a need for an internal search vehicle. Will it happen in my lifetime? I am doubtful.

There won’t be one new AI tool that can replace every application that is used – i.e. HR, GIS, Law Enforcement, Finance. Nor do I envision one AI tool that would be able to run a search through all the applications used by a department or user.

If someone was going to try and bring in internal search capabilities, it would seem like it would involve creating individual searches/agents for each use case. It would be similar to how document management systems are configured and deployed today. You figure out the documents, metadata, access, and storage location and then you deploy the solution out department by department.

This may be an area that does not have AI adoption potential. I haven’t read about any of the leading document management vendors announcing AI-related updates. Nor have I read about new companies releasing something comparable to what document management and records management solutions offer.

It feels like a good place to be, from a job security standpoint.

I will keep you posted if I hear about any new solutions.

Visit our ongoing collection on AI Adoption and Usage.