01The model
What are the four TIME outcomes?
TIME is a portfolio model popularized by Gartner. Plot each application by business value (high or low) against technical fitness (high or low). Where it lands sets its outcome.
Tolerate
Low value, good fit. Leave it and spend nothing extra.
Invest
High value, good fit. Fund new features and keep it current.
Migrate
High value, poor fit. This is the modernization queue: fix the platform or the code.
Eliminate
Low value, poor fit. Retire it or replace it with something existing.
02Scoring inputs
What goes into the two scores?
Score each input from 1 (poor) to 5 (strong). Average within each axis, or weight the inputs that matter most to you and keep the weights fixed across the estate.
| Axis | Input | Source |
|---|---|---|
| Business value | Revenue or compliance dependence | Application owner |
| Business value | Number and importance of users | Usage logs, owner |
| Business value | Rate of planned change | Roadmap, ticket history |
| Technical fitness | Runtime and framework support | Build files, dependency tree |
| Technical fitness | Test safety net | Coverage and pass rate |
| Technical fitness | Platform lock-in and license cost | Code search, contracts |
| Technical fitness | Open security findings | Composition scan |
| Technical fitness | Operating cost and incident rate | Finance, incident records |
03Worked example
How does the scoring play out?
The five applications below are illustrative, not a client. Value and fitness are averages of the inputs above. The cut line is 3.
| Application | Value | Fitness | Outcome | Why |
|---|---|---|---|---|
| Order entry | 4.7 | 2.0 | Migrate | Revenue critical, unsupported server, no tests |
| Customer portal | 4.2 | 4.0 | Invest | Current stack, tested, changes often |
| Internal expense tool | 1.8 | 3.8 | Tolerate | Few users, stable, cheap to run |
| Old reporting batch | 1.5 | 1.7 | Eliminate | Output duplicated by a newer report |
| Policy admin | 4.5 | 2.8 | Migrate | High value; fitness just under the line |
04Sequencing
How do you order the Migrate list?
Within Migrate, sort by four questions, in this order.
- 01
Value at risk
Which applications cost the most if they fail or stop being supported?
- 02
Shared dependencies
Which fixes unlock others, such as a common library or a shared server?
- 03
Cost of delay
Which license or support dates are fixed?
- 04
Effort and confidence
Which are cheap to move and well tested? Take them early to build momentum.
05Pitfalls
Where does rationalization go wrong?
- Scoring by opinion. Every input needs a named source, or the list will not survive an argument.
- Skipping Eliminate. Retiring an application is the cheapest modernization there is.
- Treating the matrix as final. An application can move quadrants after an upgrade, so rescore each year.
- Mixing inventory and assessment. Inventory lists what exists. Assessment scores it. Do not wait for a perfect inventory to start scoring the top tier.
06Related term
Portfolio assessment vs rationalization: are they different?
Application portfolio assessment gathers and scores the data. Rationalization acts on it, deciding each application's outcome. Most programs do both in one pass. After placement, each Migrate item needs a further choice of how: see the rehost, replatform, refactor and rewrite guide.
A sample assessment report that shows this scoring on a real codebase is pending. We are running this test; results will be published here.
07How Ace8 helps
Get your own scored portfolio
Ace8 is the AI-assisted application modernization practice of AceMQ.
The fixed-price modernization assessment scores your applications on inputs like those above, places each one, and returns a phased plan. For examples of what the technical inputs find, see technical debt examples.