How the IBM WebSphere Connector for Search Helps Businesses Get More Done

Searching for information at work is one of those problems nobody talks about seriously — but the numbers are ugly. IDC’s Information: The Lifeblood of the Enterprise found knowledge workers burn about 2.5 hours a day just looking for things. That’s 30% of the workday, gone. McKinsey’s version of the same finding: hire five people, and one of them is essentially just searching for answers all day, contributing nothing. For a company with 1,000 knowledge workers, IDC put the weekly cost of that at $48,000.

The 2025 Enterprise Search Survey by Slite updated those numbers. Average worker: 3.2 hours a week searching. Per year, per person, that’s 166 hours. Take a team of 50, and that adds up to 8,320 hours lost annually — just to finding things that already exist somewhere.

A big chunk of that “somewhere” is locked inside enterprise middleware. WebSphere is one of the main places it hides.

So what does the connector actually do?

The IBM WebSphere Connector for search pulls WebSphere data — documents, records, application content, metadata — into enterprise search platforms like SharePoint, Azure AI Search, and Elasticsearch. Once indexed, that content shows up in search results alongside everything else, rather than sitting invisible inside WebSphere, where most employees would never think to look for it.

Without the connector, WebSphere content is essentially a dead zone for search. Employees either don’t know the information exists or spend time duplicating work that’s already been done because they couldn’t find it.

How it’s set up

A few things worth knowing about the technical setup:

• No server-side installation. The connector is agentless — it only needs read access to WebSphere. Nothing gets installed on the WebSphere servers themselves.
• Two crawl types. Full crawls index everything. Incremental crawls pick up only what’s changed since the last crawl, so the index stays current without reprocessing the whole environment each time.
• Security stays in place. WebSphere’s existing permission model carries through into search results. Users only see content they’re already authorized to access — nothing gets exposed that shouldn’t be.
• Low system impact. BA Insight’s documentation describes it as high throughput with minimal impact on the production system, meaning WebSphere’s other operations aren’t disrupted during crawling.

The time problem is actually being solved

IDC’s research found 60% of company executives said their employees were being held back specifically because they couldn’t find information fast enough. Not a technology problem in the abstract — a finding-things problem. McKinsey’s data shows 20% of a typical digital worker’s day goes to searching for internal information alone, separate from time spent in email.

When WebSphere content gets pulled into unified search, employees stop hitting that dead zone. The Slite 2025 survey numbers show the scale of what’s recoverable: 166 hours per person per year currently lost to redundant searching. The connector doesn’t fix all of that, but it closes one of the bigger gaps — content that was always there but never findable.

A note on the wider WebSphere platform numbers

The connector sits inside a platform that carries its own performance track record. Two Forrester studies commissioned by IBM put numbers on it:

• The Total Economic Impact of WebSphere Liberty (Forrester Consulting, 2018): developer productivity up 25%, infrastructure utilization improved by 30%
• The Total Economic Impact of Migrating From Open Source Application Servers to IBM WAS Liberty (Forrester Consulting, February 2016): 122% ROI for organizations migrating from open-source Java EE servers, with investment payback in 16 months

These apply to the Liberty runtime, not the search connector specifically. But they give context for the broader environment the connector operates within.