Superposition one plan, from requirement to decision Request access

The best engineering project management software

It depends which of five jobs you need done. Here is how to tell which one, and where we fit.

Which question are you asking?

General work management

“What is everyone doing this week?”

Boards, lists and assignees. Quick to set up, but they don't track dependencies, cost or the design itself.

Critical-path scheduling

“When does this actually finish?”

Dependencies, float and baselines, in tools that have done this well for decades. Usually one planner runs it, and the engineers rarely see it.

Cost control and earned value

“Are we going to overrun?”

Control accounts, burdened rates and variance. Often run from the accounting system, so the numbers are a month old.

PLM and MBSE

“What is this made of?”

Parts, configurations, requirements and architecture. The master record of the design, but it can't tell you what a change costs.

Portfolio management

“Which program should we fund?”

Totals across many programs, for the people who decide where money goes. Good for the big picture, but too high-level for one engineering change.

The gap

“All of them, connected.”

Each kind of tool above does its own job well. Almost nothing connects them, and the hard questions, like what a slip costs, need all of them at once.

Where it hurts

Connecting them by hand takes weeks

A long-lead casting slips three weeks. The scheduler has the date, the cost system has the money and the PLM has the part, so someone builds the answer in a spreadsheet over about two weeks.

What we do about it
The coffee maker program as a Gantt chart: the design gates across the top, the critical chain running through the moulding tool, and late work marked in red.

Six questions to ask before a demo

  1. Start from what you can't answer today

    Start there, before any feature list. If you can already say what a change costs and how long it delays you, buy a tool for coordination or scheduling instead.

  2. Check what it needs you to abandon

    If a tool needs your schedule, cost data and parts moved into it, you are buying a migration. Migrations are where these projects often fail.

  3. Ask where the numbers come from

    If nobody can show how a number was calculated, it won't survive a review. Ask to see the same number calculated twice from the same inputs, with the same result.

  4. Decide about AI before the demo

    Ask what the assistant does when it doesn't know. If it always has an answer, some of those answers are guesses.

  5. Establish where the data has to live

    For defense, nuclear and much of aerospace, this comes first. Ask whether the version that runs without internet has every feature, or only some.

  6. Find out what happens when you leave

    Ask for an export before you sign. If no other tool can open it, you can't leave.

Where we fit, and where we don't

Don't buy this if

  • You need work management for a whole company.
  • Your program is one person and a list.
  • You need it in production this quarter. The capabilities page says what is still in progress.
  • You want a portfolio roll-up across forty programs.

It is worth a conversation if

  • Your schedule, budget and design data disagree.
  • Somebody senior asks what a change costs and hears back in two weeks.
  • Your data cannot leave the building.

Why no products are named

Feature lists change every quarter, so any comparison table would soon be wrong. If you want our view of a tool you already use, ask us.

Still not sure which you need?

Tell us what you use today. If one of the other four kinds of tool fits you better, we'll say so.