You can run this software project assessment without reading code. It covers six areas of three statements each: Visibility, Delivery, Scope, Ownership, Quality and Communication.
Points and area scores
- Points: Yes 3, Mostly 2, Sometimes 1, No 0.
- Don’t know: 0 on the Ownership statements (Q10 to Q12), 1 everywhere else, and always listed as a blind spot.
- Area score = points ÷ 9 × 100, shown once all three of its statements are answered.
Overall score and bands
Overall score = (1.5 × Ownership + 1.25 × Visibility + 1.25 × Delivery + Scope + Quality + Communication) ÷ 7, rounded to a whole number and shown once all 18 statements are answered. Ownership weighs most because without the code and accounts, no other team can take over. Visibility and Delivery show whether progress is real.
Red flags that override the score
Each red flag appears as soon as its answer is given, before all 18 are done.
Two overrides then run, in this order. First, if the code flag (Q10), the accounts flag (Q11) or the missed-dates flag (Q4 and Q5) is raised, or any two flags are, a Healthy or Watch closely band becomes At risk. Second, a stalled project with a rounded score under 60 becomes Needs intervention. The results panel then shows what the score alone would have given.
How the top 5 priority actions are chosen
Each answered statement scoring 1 or less gets a priority of area weight × (3 − points). The five highest are shown, ties in question order, each with why it matters and one question to ask this week, such as “When did we last restore a backup, and how long did it take?” The meeting-questions list includes every low-scoring statement, not just five.
Worked example (hypothetical, not client data)
Example (hypothetical): a project in development with an agency, answered like this.
Overall = (1.5 × 55.6 + 1.25 × 77.8 + 1.25 × 55.6 + 66.7 + 55.6 + 88.9) ÷ 7 = 65.9, shown as 66. That’s Watch closely by score, but “Don’t know” on Q11 raises the accounts flag, so the band is At risk.
Priorities: Q11 = 1.5 × 3 = 4.5, Q6 = 1.25 × 2 = 2.5, then Q8 and Q14 at 1 × 2 = 2, with Q8 first by question order. Had the project stalled, the stalled flag would be added, but 66 isn’t under 60, so it would stay At risk.
Assumptions and limits
- It scores signs you can see from the outside. It isn’t a code review.
- The weights and thresholds are our judgment, not an industry standard.
- There’s no “not applicable” answer. If you’re still planning, answer for what’s been agreed or promised so far.
- “Who builds it?” only changes who the “Ask … this week” label names (your agency, your freelancer, your team) and is noted in the summary you can send. It never changes the score.
- The “Nobody else could run it” and “No working software” flags need a No; “Don’t know” doesn’t raise them. “Don’t know” on Q4 and Q5 counts as 1, so it can raise the missed-dates flag. The two stage flags need a stage chosen.