Implicit
Case Studies / Software Support Zendesk Integration

From three knowledge bases to one source agents trust

A global SaaS company formed through a three-way merger used Implicit's walled-garden AI, connected directly to its Zendesk knowledge bases, to turn fragmented, multilingual support knowledge into cited, source-backed answers its agents could trust.

When three companies merge, their knowledge doesn't merge with them. Built from three businesses across three continents, this global SaaS company ran three legacy support systems at once, with answers scattered across separate Zendesk knowledge bases and an internal wiki nobody owned. It turned to Implicit to consolidate that knowledge into one governed, source-backed layer, without handing an AI content it couldn't trust.

The knowledge problem

The external knowledge was solid. Each business ran its own Zendesk knowledge base, kept current by active technical writers. The internal knowledge was the problem: years of institutional know-how sat in Confluence with no owner, no update cadence, and no reliable way to tell what could still be trusted.

As support shifted from self-serve toward high-touch service, agent enablement moved front and center. Zendesk's native search couldn't handle the detailed, multi-article questions agents asked most, and early experiments with generic AI failed on the thing that mattered: answers agents couldn't trace, and therefore couldn't trust.

3

Legacy systems to unify

Three merged businesses, each with its own products, teams, and documentation, consolidating into one platform.

600+

Articles in the first Implicit workspace

External help content alone in the initial pilot scope, with internal content still being scoped.

6

Languages in use

Support runs across multiple languages globally, so answers have to hold up across all of them.

One governed, source-backed layer

Rather than point a chatbot at every file it could find, the company built one curated knowledge layer per product, feeding it only verified content. Implicit's GraphRAG engine structures that content into a queryable graph where every answer traces back to its source.

Back to sources Implicit

Source · Support KB

SSO troubleshooting

Cited passage [1] If a directory sync fails, SSO is disabled automatically. Re-run the sync from Admin → SSO, then re-enable SSO. Active sessions remain valid.
A customer can't log in with SSO after last night's directory sync failed. How do I fix it?

Answer

Re-run the directory sync from Admin → SSO, then toggle SSO back on.1 Active agent sessions stay signed in, so no one is logged out mid-ticket.

If the sync fails again, confirm the SCIM token hasn't expired before retrying.2

1 SSO troubleshooting · Support KB
2 Directory sync setup · Support KB
Suggest follow-up questions Turn into article
Every answer is assembled from cited passages in the customer's own knowledge base. Illustrative UI.

From ingestion to answer, the flow runs through a governed pipeline:

  1. 1

    Connect the real sources

    Implicit ingests each product's Zendesk help-center content directly from its root URL and pulls in internal material from Confluence and Google Drive, unifying knowledge that used to live in separate silos.

  2. 2

    Isolate by product

    Separate workspaces keep each product's knowledge cleanly walled off, so the AI serving one legacy product never pulls from another's docs, and content can be added incrementally as it's verified.

  3. 3

    Structure it into a graph

    GraphRAG auto-generates a taxonomy of concepts and keywords and maps the relationships between them, so retrieval reaches the right passage even when the answer spans multiple articles or sits deep inside a long one.

  4. 4

    Keep it live

    Sources sync automatically from the Zendesk backend on a daily cadence, so the knowledge base stays current as technical writers ship updates, with no manual re-upload.

  5. 5

    Answer with citations

    Every response links back to the exact source, with the AI's reasoning visible, and when the material can't support an answer, the system says so rather than inventing one.

  6. 6

    Close the feedback loop

    A thumbs-down prompts agents for context, then routes the signal by type, bad source data versus a knowledge gap, feeding gap detection, drafting, and an approval workflow before anything re-enters the knowledge base.

  7. 7

    Deliver it where agents already work

    A resizable Chrome sidecar brings the knowledge base into any browser tab, an MCP layer exposes the same knowledge to AI agents, and in-context translation serves the multilingual team.

Native search vs. governed retrieval

Before Implicit, agents relied on Zendesk's native search. It handled simple lookups but fell short on the detailed, multi-article questions that filled the queue. That gap in how each approach finds an answer is what pushed the team to look for something better.

Native knowledge-base search

One article at a time

Keyword-matched a single article, and struggled with detailed questions that spanned several or lived deep inside one.

Implicit governed retrieval

Across the full curated set

Graph and vector retrieval across the whole curated set, every answer cited and source-linked, at high accuracy with minimal tuning.

Key capabilities

Walled-garden curation

The AI only ever sees content the team has verified. Knowledge is added deliberately, workspace by workspace, instead of scraping entire sites and hoping for the best.

Cited answers and visible reasoning

Every answer links to its source and shows how it was reached, so agents can validate it in seconds. This was the feature that turned AI from a risk into something the team would actually rely on.

Multilingual delivery

In-context translation serves a team working across five to six languages, so agents get trusted answers in the language they work in.

Governed access and roles

Role-based permissions and per-workspace scoping give the knowledge team control over who can add, approve, and edit, with approval workflows on the roadmap.

A Chrome Extension, redesigned together

Support agents and CSMs live in their browsers, moving between Zendesk, help-center articles, and internal docs all day. Working closely with the customer support team, we completely redesigned Implicit's Chrome Extension around how CSMs actually work. A resizable sidecar puts trusted, cited answers next to whatever page is open. It allows users to add content to their knowledge base, ask questions about the webpage they are on, and switch between workspaces as needed. It is now included in every Implicit product edition.

support.example.com/hc/en-us/articles/re-enabling-sso
Implicit Chrome
Workspace Support KB
How do I re-enable SSO after a failed sync?

Re-run the directory sync from Admin → SSO, then toggle SSO back on. Active agent sessions stay signed in, so no one gets logged out mid-ticket.

SSO troubleshooting · Support KB [1]
Answer from this page · cited
The redesigned sidecar answers from the workspace and the page in view, every answer cited. Illustrative UI.

Answers in context

Ask about the whole workspace or just the page, or selection, an agent is looking at, without leaving the tab.

Sessions that follow

Chats sync both ways, so an agent can start in the extension and pick it up later inside Implicit.

Save as you go

Capture any page or highlight straight into the right workspace to keep the knowledge base growing.

My team practically lives in the browser, so having Implicit right there in the sidecar, next to whatever page they're on, changed how they work. We walked the Implicit team through how our CSMs actually operate, and they rebuilt the extension around it.
— Content & Knowledge Manager, global SaaS support team

A partnership, not a handoff

The results came from how the two teams worked. The customer's first dedicated content and knowledge manager shaped the rollout from inside the support org, and Implicit turned feedback around fast, often folding requests into the product within days. That tight loop is what moved the pilot from a test into something the team wanted to expand.

What gave us confidence was the speed. We would raise something and often see it in the product within days. And because every answer is cited, my agents trust what comes back instead of double-checking it themselves.
— Content & Knowledge Manager, global SaaS support team

The results

The customer now has one place to find trusted answers across a sprawling mid-merger knowledge landscape, without giving up control of how the AI will answer, and where it will pull those answers from. Instead of bouncing between multiple tools and knowledge bases, everyone and everything can pull from one source of truth: Implicit.

~98%

Answer accuracy in pilot

Reached with minimal configuration or tuning, even while the pilot team was temporarily short-staffed.

722
Live

Sources syncing live

Help-center content syncing from Zendesk and auto-updating daily in the initial workspace, so answers stay current on their own.

Daily sync volume

Answers agents can verify

Citations and visible reasoning let agents check every answer against its source, clearing the barrier that had stalled earlier AI experiments.

Built for a moving target

Daily Zendesk sync and workspace isolation let the team keep answers accurate across three legacy products while building toward one unified platform, instead of freezing knowledge in place.

Give your support team answers they can trust.