Skip to content
Septiva

Second Opinion

A better estimate begins with better assumptions.

Before approving a material implementation commitment, examine the assumptions behind the delivery model.

Purpose

The purpose is not automatically to replace the incumbent provider.

This is not

“We have already decided the current vendor is wrong.”

This is

“This commitment is important enough to challenge the assumptions before we approve it.”

The decision model

Six legitimate outcomes

Embedded in the proposal
Under reviewScope

Is the initiative broader than the evidence currently justifies?

Legitimate outcomes
  • Approve
  • Modify
  • Decompose
  • Stage
  • Supplement
  • Reconsider

A second opinion challenges the delivery model, not merely the price.

Before you approve

Before you approve the old approach, see what the new one can do.

The deliverable

Challenge the delivery assumptions. Then decide with better inputs.

Assumption review
What the proposed timeline and cost assume about team structure, sequencing, coordination overhead, and how much of the scope is genuinely novel engineering.
Model alternatives
Where an AI-native delivery model would change the shape of the work — and, just as importantly, where it would not.
Risk and reversibility
Whether the first commitment can be made smaller without weakening the outcome, and what evidence would justify continuing.
A clear recommendation
Proceed as planned, proceed with a modified model, or start smaller. If the existing plan is sound, we will say so plainly.

Illustrative Second Opinion — Executive Output

What the deliverable looks like on one page.

The example below is constructed to show the shape of the output. It is not drawn from any client engagement and contains no real scope, pricing, or outcome.

Illustrative Second Opinion — Executive Output. Constructed to show the shape of the deliverable. Not drawn from any client engagement; no real scope, pricing, or outcome.

Current commitment

Proposed scope
A full replacement platform delivered as one programme.
Delivery model
Sequential phases, discovery completed before build begins.
Timeline
Measured in quarters, with first working software late in the sequence.
Staffing structure
A large blended team sized at the start and held constant.
Integration assumptions
All systems of record integrated before any user sees value.

Assumptions under review

Staffing
Does team size reflect the engineering work, or the incumbent delivery model?
Architecture
Which commitments are needed now, and which could remain reversible?
Sequencing
Could useful capability reach a real user before the full programme?
Integration
Which dependencies are genuinely required for first value?
Timeline
How much elapsed time is engineering, and how much is coordination?
Dependency
Which elements create avoidable long-term dependence?

What Septiva may challenge

  • Whether full scope is justified before evidence exists.
  • Whether team size reflects the actual engineering work.
  • Whether some architecture decisions can remain reversible.
  • Whether useful capability can appear materially earlier.
  • Whether long-term dependency is avoidable.

Possible recommendation

  • Approve
  • Modify
  • Decompose
  • Stage
  • Supplement
  • Reconsider

Stage is highlighted only as an illustration. Approve is a legitimate outcome, and a sound plan should be confirmed as sound.

Decision logic

Current assumptionEvidence neededRecommended next move
The full platform must be committed as one programme.Whether a bounded slice can produce real usage from real staff.Stage the commitment: fund the first slice, define what would justify the rest.
Team size is fixed by the delivery model.What the first slice actually requires of engineering.Let team size follow the problem once the work is observable.
All integrations precede user value.Which integration is load-bearing for the first useful outcome.Sequence integration behind first value rather than in front of it.

The decision, not the implementation

Build vs. Buy is no longer enough.

Modern enterprise software decisions have more than two answers. Naming the right decision can be more valuable than negotiating the price of the wrong one.

  1. 01Buy

    A commercial product already solves the problem well enough.

  2. 02Build

    The problem is specific to the organization and worth owning.

  3. 03Co-Build

    The software matters, and so does the internal capability to continue it.

  4. 04Automate

    The work is a workflow problem before it is an application problem.

  5. 05Integrate

    The capability already exists in systems that do not yet talk to each other.

  6. 06Modernize

    The system is sound in intent and constrained by its implementation.

  7. 07Retire

    The system persists out of habit rather than value.

  8. 08Defer

    The evidence does not yet justify a commitment of this size.

Our objective isn’t to build more software. It’s to make better software decisions.

Read: Build vs. Buy is no longer enough

Second Opinion or Challenge Septiva?

Two different questions. Two different starting points.

Second Opinion

A specific commitment is already on the table.

A proposal, estimate, or contract is under review and a decision date is approaching. The review examines the delivery assumptions behind that document so the approval is defensible in the boardroom.

Challenge Septiva

Not yet at the proposal stage?

If you have an important initiative, backlog item, workflow, or idea but not yet a defined implementation proposal, start with Challenge Septiva. The conversation is about what would need to be true for it to move materially faster.

Challenge Septiva

Request a Second Opinion

Before approving the estimate, challenge the model behind it.