Feedback to opportunities and delivery
A request tells you what somebody experienced. An opportunity describes a problem worth solving. A roadmap item describes the work you choose to do about it. Keeping those separate lets several requests support one decision while each original need remains available to check later.
What it does
Open Items and choose Feedback inbox or Opportunities. Roadmap items keeps the existing searchable working list. You can continue creating items directly; feedback and assessment are optional.
Plan and track the work you choose to do.
Capture, review and retain what people said about your product.
Group related needs and decide which problems to address.
Feedback and opportunities stay outside stakeholder exports and the roadmap health score. Selecting an opportunity does not add work to the canvas.
When to use it
Use it when several requests compete for roadmap space, when related requests need one product decision, or when you want to explain why work was selected.
When not to use it
If you already know the work to add, use Add item. Delivery remains on the existing item, with links to your delivery tool when useful. This flow does not add task assignment, sprint planning or automatic updates from another tool.
Capture, group and decide
Capture feedback
In Items → Feedback inbox, press Capture feedback. Only the text is required. Add the source, person and affected group when known.
Review the original need
Open the feedback to read its wording, source and context. Sort on the canvas or in Check opens this same review. Create or link an opportunity, decline with a reason, or leave for later.
Group related feedback
Choose an existing opportunity or press Create opportunity. Creation reuses the original words immediately; use the pencil to edit them. The request leaves the unsorted count but remains retained. Related requests need not be duplicates.
Assess and decide
In Assess, press Add MICE if useful. Each rating has a name; What do 1–4 mean? explains the whole scale before you choose. Choose Investigate, Selected, Not now or Declined. Explain decisions beyond Investigate; Save & continue takes a Selected opportunity directly to Work.
Create or link work
Work always offers Create roadmap item. Confirm its title and workstream. If the opportunity is not selected, add a reason and press Select & create item to record both actions together. New work starts without dates or a horizon. To link work already planned, open Link existing work, search by name or number, and choose from the workstream and item hierarchy. Press Link selected item to confirm. The existing item's plan stays as it was.
Track delivery on the item
Press Open item on a linked card to update delivery status. Linking means the work is connected; Done means built and accepted. Mark it Released when it is available to customers.
Check Results after release
Return to the opportunity after a linked item is Released. Results shows the delivery statuses, each original request and the overall outcome. Released does not mark either result as achieved automatically.
For example, “find old invoices” and “filter invoices by year” can support one opportunity: help customers find previous invoices quickly. Search and date filters can be separate work items. Releasing search may address the first request while the second remains open.
MICE assessment
MICE compares estimated value relative to effort. It helps you discuss trade-offs; it does not set priority, build order, horizon or dates. Assessment confidence is separate from confidence in a roadmap item's delivery plan.
| Factor | What you assess | Scale |
|---|---|---|
| Magnitude | How many people are affected, and how often | 1: few people, occasional need. 2: several people, recurring need. 3: many people or frequent need. 4: widespread, frequent need. |
| Impact | Improvement to the user experience or business outcome | 1: minor improvement. 2: noticeable improvement. 3: significant user or business benefit. 4: substantial user or business value. |
| Confidence | Strength of evidence behind the estimates | 50%: assumptions or limited evidence. 80%: evidence with uncertainties. 100%: strong supporting evidence. |
| Effort | Design, development, testing, complexity and dependencies | 1: limited change, low complexity. 2: moderate effort. 3: significant effort or dependencies. 4: major effort or complex dependencies. Higher effort lowers the score. |
Score = Magnitude × Impact × Confidence ÷ Effort. For example, 4 × 3 × 0.8 ÷ 2 = 4.80. These are the initial working scales; custom formulas and weights are not available.
The confidence buttons show percentages; the calculation uses 0.5, 0.8 or 1. The explanations are available before selection and do not require hovering.
Use the same assessment period and similar scope across the opportunities you compare. Record that context under Affected people and assessment period. Choose MICE in the opportunity list's Sort control to put complete scores first and show the four inputs. Partial assessments remain unranked. Choose Recent to return to the ordinary list; comparison never writes a roadmap priority.
Delivery and outcome review
Open linked work to update its existing status, ownership, priority and dates. Its panel has Open related opportunity to return to the problem and evidence. New work starts unscheduled: Horizon shows No horizon yet until you choose a value. Selecting Now, Next or another horizon saves it and removes the missing-horizon warning.
Results shows how many linked items are Released and which still await release. One released item is enough to start checking its effect; the rest may still be in progress. Done is not treated as Released: it means built and accepted, possibly before customers have it. Existing reviews remain available.
With no released work, Results explains what to do next. Record an early finding lets you record evidence from a pilot or another check before release; it does not change item status or mark the opportunity successful.
In Results, open an original feedback card. Did this address the request? lets you record Addressed, Partly addressed or Still open, with what you checked. Addressed and partly addressed results require an explanation. Later checks of that request are dated and kept with the earlier findings, so changing the answer does not erase what was observed after the first release.
In Results, use Overall outcome for the result: Achieved, Partly achieved or Not achieved, with review notes. This is a separate judgment from whether every feature is released. Saving a review does not close other requests.
When the overall result is Partly achieved or Not achieved, Results asks What next? Continue planned work, try another response in Work, learn more, stop for now, or reframe the problem. The result stays visible. A new result check is dated and added to the earlier checks; an older review with no known review date is shown as such. Reframing creates a new opportunity linked to the earlier one. Use it when the problem or intended outcome has changed. Earlier work and feedback remain on the original opportunity; nothing is moved or closed automatically.
Finding records as the lists grow
Feedback inbox and Opportunities use the same Search, Sort and Filter controls as Roadmap items. Search in Feedback includes the original wording, source, person, affected group, decline reason and linked opportunity. Filter by Waiting, Linked or Declined, or narrow by review result, source, capture age and linked opportunity. Opportunities can be filtered by decision, overall result, owner, linked work and whether an assessment is complete. The result count shows how much of the list you are looking at. Filters stay in place while you move between these views during the current session; use Clear all filters to show everything again.
Choose an assessment method
In Config → Opportunity assessment, enable the methods this roadmap uses and select a default comparison method. Existing MICE assessments stay saved if you hide that method. In Assess, switch between the enabled methods; each keeps its own inputs. The Sort control compares opportunities using one named method at a time. Incomplete assessments stay unranked, and scores never set roadmap priority, build order or dates.
| Method | Use it when | Inputs |
|---|---|---|
| MICE | Affected reach is still an estimate | Magnitude, impact, confidence, effort on the existing scales |
| RICE | You can estimate reach over a stated period | Reach count, period/unit, impact, confidence, person-months of effort |
| Value vs effort | Evidence is thin and a rough comparison is enough | Value and effort from 1 to 4; shown as a quadrant, with no decimal score |
MICE magnitude is a rating; RICE reach is a count for a named period. The app does not convert one to the other. Use notes to say what supports the estimate. An edited assessment receives the date when you save it. Existing assessments without a known date remain undated.
Best use cases
- Preparing a roadmap review with evidence behind the choices.
- Grouping related requests without losing their different requirements.
- Checking what actually improved after a release.
Common mistakes
- Counting feedback records as the magnitude rating. Several records can come from the same people; assess the affected population and frequency instead.
- Treating the highest score as an instruction to build. Dependencies, obligations and capacity still matter.
- Treating Selected as a delivery promise. Schedule the linked work explicitly.
- Marking every request addressed when one feature is released. Check each need.
Related guide pages
Availability
Free, on every plan. Records are saved with the workspace and its backups. Roadmap transfer JSON includes feedback, opportunities and their links. Older app versions do not offer the workflow; use the updated app when importing or reviewing these records.
The delivery brief still has a job
On linked work, the former handover section is called Delivery context. It shows the linked opportunity and its outcome and measure. Item-specific outcome and measure are folded away; scope remains with the item. Existing notes are retained. Fill missing brief details copies only empty fields, so it cannot overwrite what you have already written.
New linked work carries the problem, outcome and measure into its brief. These are a starting copy; later opportunity edits do not silently rewrite an item's brief. The brief also quotes its linked original feedback, includes the opportunity's decision and reason, and labels MICE evidence confidence separately from delivery confidence. Items created directly keep Handover readiness and need no opportunity.