Later Roadmaps

Gantt charts are where product strategy goes to die. They give the illusion of precision — neat bars stretching across months, dependencies drawn with arrows, milestones pinned to specific Tuesdays. The problem is that none of it is real. You do not know what you will learn in three months, what the market will do, or which assumption will turn out to be catastrophically wrong. A Gantt chart cannot encode uncertainty. A Now/Next/Later roadmap can.

The Core Idea

Now/Next/Later is a time-horizon framework that replaces fixed dates with confidence levels. It divides your roadmap into three buckets:

Now   — Currently in progress or starting this sprint/week.
        High confidence. Clearly defined. Teams assigned.
        This is what we are building right now.

Next  — Coming up after current work completes.
        Medium confidence. Themes and problems identified,
        but solutions may not be fully scoped.
        Roughly "this quarter" but no specific dates.

Later — On the horizon. Aligned with vision and strategy.
        Low confidence. Directional, not detailed.
        May shift dramatically based on what we learn.

The framework was popularized by Janna Bastow, co-founder of ProdPad, as a direct response to the dysfunction caused by date-driven roadmaps in product teams.

Why It Works

It Matches Reality

The further out you look, the less you know. Now/Next/Later reflects this honestly. A feature you are building today has gone through discovery, design, and scoping — you can speak about it with confidence. Something six months out is a guess dressed in a spreadsheet. Treating both with the same level of specificity is dishonest.

It Forces Prioritization

With only three buckets and no infinite timeline to push things into, you have to make choices. If something is not in Now or Next, it goes in Later — which means it might never happen, and that is fine. The roadmap becomes a prioritization tool, not a feature parking lot.

It Reduces Deadline Theater

When every item has a date, every item becomes a deadline. Deadlines create pressure to ship regardless of quality. Teams cut corners, skip testing, and ship half-baked features to hit an arbitrary date that someone typed into a spreadsheet six months ago. Now/Next/Later removes the date pressure from items that do not need it.

How to Build One

Step 1: Define "Now"

List everything currently in active development. These items should have:

- A clear problem statement
- Defined acceptance criteria
- An assigned team or owner
- A rough estimate (days or sprints, not months)

If your Now column has more than 3-5 items per team, you are overcommitting. Teams cannot make meaningful progress on eight things simultaneously. Cut the list.

Step 2: Define "Next"

These are the problems you plan to tackle after the current work. Frame them as problems or themes, not solutions:

Weak Next items:
  "Build Slack integration"
  "Add CSV export"
  "Redesign settings page"

Strong Next items:
  "Reduce time-to-first-value for new users (activation is 30%)"
  "Enable team collaboration on shared resources"
  "Address top 3 churn reasons from exit surveys"

Notice that strong Next items define the problem, not the feature. This leaves room for discovery. You might find that the best way to improve activation is not a Slack integration — it might be simplifying the onboarding flow.

Step 3: Define "Later"

Later is strategic and directional. It connects to your product vision but stays deliberately vague:

Later examples:
  "Expand into the enterprise segment"
  "Build a platform/API for third-party integrations"
  "Explore AI-assisted workflows"

Later items are not commitments. They are directions you are considering. They might move to Next as you learn more, or they might get dropped entirely.

Step 4: Review Regularly

A Now/Next/Later roadmap is a living document. Review it at least every two weeks:

Review cadence:
  - Weekly: Update "Now" based on progress and blockers
  - Biweekly: Reassess "Next" based on new learnings
  - Monthly: Revisit "Later" based on strategy changes
  - Quarterly: Full roadmap refresh with stakeholders

Items flow between buckets. A Next item might jump to Now because of a customer emergency. A Now item might drop back to Next because discovery revealed unexpected complexity. This movement is healthy — it means you are responding to reality.

Communicating Uncertainty Without Looking Incompetent

This is the hard part. Leadership wants dates. Sales wants commitments. Customers want to know when their feature is coming. Saying "it is in the Later bucket" feels evasive. Here is how to handle it.

Use Confidence Language

Instead of:                     Say:
"I don't know when"             "We're targeting Q3, with medium confidence"
"It depends"                    "It depends on the outcome of X, which
                                 we'll know by end of month"
"We can't commit to that"       "We can commit to the problem, not the
                                 solution — here's what we're exploring"

Tie Uncertainty to Reasons

People accept uncertainty when they understand why it exists. "We do not have a date" is frustrating. "We do not have a date because we are still running user interviews to validate the approach, and we would rather build the right thing than build the wrong thing fast" is reasonable.

Show Your Track Record

If you delivered your last three Now items on time and within scope, people will trust your Next estimates. Build credibility through consistent delivery in the near term, and stakeholders will give you slack on the long term.

Provide Context, Not Just Status

Bad roadmap update:
  "Feature X: In progress"

Good roadmap update:
  "Feature X: In progress. Discovery complete. 2 of 4 milestones
   shipped. On track for end of sprint 14. Key risk: third-party
   API reliability — we have a fallback plan if it doesn't work."

Real-World Example: A B2B SaaS Company

A mid-stage B2B SaaS company had a Gantt chart roadmap that went out 18 months. The CEO presented it to the board every quarter. The problem: they had never once shipped according to that 18-month plan. Every quarter, the Gantt chart was rebuilt from scratch. It was fiction presented as fact.

They switched to Now/Next/Later:

Now (this sprint):
  - Fix SSO login failures (blocking 3 enterprise deals)
  - Ship bulk import for CSV data (top support request)
  - Complete API rate limiting (security audit requirement)

Next (this quarter):
  - Improve onboarding activation (currently 28%, target 45%)
  - Reduce report generation time (p95 is 12 seconds, target 3)
  - Add role-based permissions (enterprise requirement pattern)

Later (future quarters):
  - Self-serve enterprise tier
  - Mobile companion app
  - Marketplace for third-party integrations

The CEO initially resisted. "The board needs dates." They compromised: Now items had sprint-level estimates, Next items had quarter-level ranges, and Later items had no dates. After two quarters of delivering Now items consistently, the board stopped asking for 18-month timelines.

Stakeholder-Specific Views

Different audiences need different levels of detail from the same roadmap. A single Now/Next/Later board serves as the source of truth, but how you present it changes by audience.

For the board / investors:
  Show strategic themes in Next and Later.
  Show delivered outcomes in Now (not feature names).
  Focus on metrics: "We improved activation from 28% to 43%."

For sales:
  Show Now items with expected delivery windows.
  Show Next items framed as "problems we're solving for customers."
  Provide talking points, not feature promises.

For engineering:
  Show Now items with full detail: scope, milestones, owners.
  Show Next items with problem statements and discovery status.
  Later is context for why current work matters.

For customers:
  Show Now items relevant to their use case.
  Frame Next as "areas we're investing in."
  Never show Later — it creates expectations you cannot meet.

The roadmap is one document with multiple presentations. This is not spin — it is appropriate communication for different audiences with different needs and different levels of context.

How It Compares to Other Approaches

Framework          Strengths                     Weaknesses
---------------------------------------------------------------------------
Gantt Chart        Visual, date-specific         False precision, brittle
Theme-based        Strategic alignment           Can be too vague to execute
Opportunity Tree   Discovery-focused             Complex, learning curve
Now/Next/Later     Honest, simple, adaptable     Requires trust from stakeholders
RICE-scored list   Quantified priority            Scores create false objectivity

Now/Next/Later is not the only valid approach, but it is the best default for product teams that need to communicate direction without committing to fiction.

Common Pitfalls

Treating Now/Next/Later as a backlog — it is not a list of everything you might build. It is a curated view of your top priorities. If your Later column has 40 items, it is a backlog wearing a roadmap costume.
Never moving things out — items should leave the roadmap, not just accumulate. If something has been in Later for three quarters and nobody has advocated for it, remove it.
Making Now too big — if Now has 15 items, you have not prioritized. Now should reflect what teams can actually deliver in the current cycle.
Using Later as a graveyard for rejected ideas — Later should contain strategically relevant directions, not ideas you are too polite to say no to. Say no explicitly instead.
Skipping the review cadence — a roadmap that is not reviewed regularly becomes stale. A stale roadmap is worse than no roadmap because people make decisions based on outdated information.
Letting sales dictate Now — every deal looks urgent. If you let individual sales requests jump straight to Now without going through prioritization, your roadmap is just a sales request queue.

Key Takeaways

Now/Next/Later replaces false precision with honest confidence levels, matching the reality that uncertainty increases over time.
Frame Next and Later items as problems and themes, not features — this preserves room for discovery and prevents premature solution commitment.
Roadmaps are living documents that should be reviewed regularly, with items flowing between buckets based on new information.
Communicate uncertainty by pairing it with reasons, context, and a track record of near-term delivery.
The goal of a roadmap is alignment, not a contract. If your roadmap is never wrong, you are either not looking far enough ahead or not being honest about what you know.