Back to case studies
Customer Success

From five tabs to one customer context workspace

A 26-person B2B SaaS team had five CSMs toggling across CRM, calls, support tickets, and product usage to prepare for every account conversation. We unified the evidence into one customer context workspace, embedded where the team already worked.

Timeline:8 weeks · discovery to production handover
Delivery:Production handover
Analytics dashboard showing connected customer data

Project snapshot

The small-team operating context

The delivery and value signals below are anonymized and rounded so the operating scale is visible without exposing the client.

Client context

B2B SaaS company · 26 employees · 5 CSMs · ~160 customer accounts

Delivery team

2 engineers + 1 product lead with the Head of Customer Success

Sponsor

Head of Customer Success

Account preparation

Hours → seconds

Evidence-linked account view

Project-reported

Source switching

5 tools → 1 workspace

CRM, calls, tickets, and usage

Project-reported

Capacity pathway

~20 hrs/week

Directional CSM prep capacity

Directional model

Production signal

  • CSMs used the account brief before renewal, risk, and expansion conversations.
  • Reviewers rejected unsupported risk signals instead of carrying them into customer-facing work.
  • The team kept approved follow-ups connected to the account record rather than unmanaged notes.

What we learned

  • Define the account identity rules before generating summaries.
  • Citations and freshness indicators matter as much as fluent prose.
  • Keep customer-facing actions behind a CSM approval step.
Delivered: Governed account context layer · Embedded customer workspace · Review workflow, runbook, and adoption instrumentation

Actual process teardown

How the operating process changed

The important change is not a model in isolation. It is the full operating path around the model: inputs, orchestration, controls, review, and the system of record.

The challenge

Five CSMs were responsible for roughly 160 customer accounts. Before renewals and risk reviews, each person manually stitched together recent calls, open tickets, product adoption, and commercial context from five tools. The work was slow, repetitive, and easy to leave incomplete.

The production response

We created a governed account context layer, added evidence-linked summaries and risk signals, and embedded the review loop in the existing workspace. The copilot drafts the account story; the CSM validates it and owns the next action.

Before · manual baseline

  1. 01

    Open five systems

    CSM

    CSMs gathered CRM, call intelligence, support tickets, product activity, and commercial context from separate tools.

  2. 02

    Build the account story

    CSM

    The team manually assembled recent events, risks, unresolved issues, and open follow-ups.

  3. 03

    Prepare for the meeting

    CSM

    Important context was summarized in notes, often duplicated, and sometimes went stale before the conversation.

  4. 04

    Decide what to do next

    CSM

    Risk and expansion signals depended on individual judgment, memory, and whichever system was checked last.

After · AI-assisted production

  1. 01

    Sync account context

    System

    CRM, calls, tickets, product events, and commercial context are joined into one governed account view.

  2. 02

    Summarize and surface signals

    AI

    AI produces an evidence-linked account brief, highlights changes, and shows the source and freshness of each signal.

  3. 03

    Review the brief

    Human reviewer

    CSMs validate the summary, add context, and reject unsupported risks before using it in a customer conversation.

  4. 04

    Act in the workspace

    CSM + system

    Approved follow-ups and next actions stay connected to the account record and remain visible to the team.

Phase-by-phase delivery

What was delivered, in order

Each phase turns the process teardown into a concrete production deliverable, with a clear value created before the next layer is added.

  1. 01Delivery phase

    Define the account context

    Identify the questions CSMs repeatedly answer before a renewal, risk, or expansion conversation.

    Delivered

    • Account-prep workflow map
    • Source inventory across CRM, calls, tickets, and usage
    • Identity and account-matching rules

    Value created

    The product is shaped around decisions CSMs need to make, not around a generic data dump.

  2. 02Delivery phase

    Unify the evidence layer

    Create one governed account view from systems that previously required manual cross-checking.

    Delivered

    • Joined account context model
    • Event freshness and provenance rules
    • Role-aware access to customer data

    Value created

    CSMs get a consistent source of context without copying sensitive data into unmanaged notes.

  3. 03Delivery phase

    Add the copilot loop

    Generate a useful brief while preserving citations, reviewer judgment, and approved actions.

    Delivered

    • Evidence-linked account summary
    • Risk and expansion signal detection
    • Draft next actions with confidence checks

    Value created

    Preparation becomes faster while the CSM remains accountable for what is communicated or changed.

  4. 04Delivery phase

    Embed and measure adoption

    Put the workflow inside the existing workspace and observe whether it changes preparation behavior.

    Delivered

    • Embedded account workspace
    • Review and correction feedback loop
    • Usage and quality instrumentation

    Value created

    The transformation compounds through adoption rather than becoming a separate dashboard people must remember to open.

How to read this teardown: the delivery sequence combines the existing public-safe project narrative with a practical view of the production work around it. It does not add client names, confidential implementation details, or unverified performance claims.

Measured and modeled impact

Separate the outcome from the value hypothesis

Project-reported outcomes are shown alongside directional models. Public benchmark figures are kept in their own evidence section below.

Account preparation

Project-reported

Hours → seconds

Before

Hours across five tools

After

Seconds to open the account view

Interpretation boundary

Anonymized and rounded project-reported outcome supplied for this case study; independently unverified in the public view.

Value model
Directional model

Retention and expansion value

For a small SaaS team, the return comes from giving five CSMs more usable account coverage without immediately adding another customer-success hire.

Model

Reviewed at-risk ARR influenced × modeled retention lift + expansion opportunities influenced × conversion value − workflow operating cost

Assumptions to validate

  • The ~20-hour weekly capacity estimate is validated through time sampling.
  • CSMs use the saved time on observable renewal, risk, or expansion actions.
  • ARR, retention lift, conversion value, and hiring cost are supplied by the revenue team.

ROI scorecard

Value across time, money, resource, output, and error

Each teardown shows the practical ROI lens for the transformation. Existing project signals stay labeled; financial impact remains a transparent model until baseline volumes, costs, and quality data are verified.

Time
Project-reported

Hours → seconds

Account preparation compresses from cross-tool research to an evidence-linked account view.

Money
Directional model

Protect renewal value

Directional model: connect reviewed risk actions to at-risk ARR, renewal coverage, and expansion opportunities.

Resource
Directional model

~20 hrs/week potential

For five CSMs, reducing repeat preparation by roughly 45 minutes per account creates capacity for customer conversations.

Output
Directional model

Broader account coverage

A shorter preparation loop gives the same team more room for renewal, risk, and expansion work.

Error
Directional model

Fewer context gaps

Joined records, citations, freshness indicators, and reviewer checks reduce unsupported account conclusions.

ROI should be finalized against the client baseline: loaded labor cost, transaction volume, unit economics, throughput, and measured error or denial rates. The directional model is a value pathway—not an invented client result.

System blueprint

The production system around the AI

The tool list shows what each component is responsible for across the application, data, AI, integration, orchestration, and governance layers.

Tools used

Application

TypeScript

Account context, review states, and approved action handling

Workspace

Next.js

Embedded customer-success workspace and account brief UI

Data + context

Snowflake

Joined customer context, event history, and governed analytics

CRM integration

Salesforce API

Account, opportunity, renewal, and approved follow-up records

Integration

Source connectors

Call intelligence, support tickets, and product-usage events

Governance

Identity + freshness rules

Account matching, provenance, access, and stale-context checks

Orchestration

  1. 1CRM and event sync
  2. 2Identity and account matching
  3. 3Evidence retrieval
  4. 4Summary generation
  5. 5Confidence and citation check
  6. 6CSM review
  7. 7Approved CRM action

In-house data and boundary

The system is grounded in the client’s existing operational data rather than an unbounded external corpus.

CRM account and opportunity recordsCall transcripts and meeting notesSupport tickets and resolution historyProduct usage and adoption events

Human in the loop

  1. 1Validate account summaries before external use.
  2. 2Confirm risk signals and customer-facing actions.
  3. 3Correct account matching, freshness, or missing context.
  4. 4Escalate sensitive or ambiguous account situations; no customer-facing action is sent automatically.
AI capabilities
Account summarizationConversation and ticket synthesisRisk-signal detectionRecommended next-action drafting

Public benchmark evidence

What the wider market is showing

These sources support the transformation pattern. They are not client-specific outcomes and should not be read as such.

Public benchmark

Deloitte — AI in Post-Sales Customer Experience

Deloitte identifies account summaries, meeting-note production, and support-ticket synthesis as practical AI opportunities for customer success.

Why it matters here

Supports the use of AI to reduce preparation effort while leaving customer judgment and action with the CSM.

External research provides context and benchmarks. It does not verify the anonymized client, implementation, or project-specific results described above.

Names, logos, and specific numbers are redacted in this public view. Full references are available under NDA on request.

Build the next one

Have a workflow that needs a closer look?

Start with the friction you already see. We will help you determine what is worth solving and how to move it safely into production.

Assess my workflow Talk to us