The Marketing Cloud kickoff always hits the same wall. The client has 20 use cases they want ("welcome series, abandoned cart, loyalty emails, personalized recommendations, re-engagement, birthday, VIP, SMS, push..."). The budget funds 4-5 of them this quarter. How do you pick?
After enough engagements we now use a 3-column scoring framework. It's not fancy. It works because it forces a conversation about trade-offs that the client otherwise won't have with themselves.
The framework
Score each use case 1-5 on three dimensions:
| Criterion | Score 1 | Score 5 |
|---|---|---|
| Business impact | Nice to have | Directly drives revenue or retention |
| Customer impact | Nobody would notice if we skipped it | Visibly improves the customer experience |
| Effort | Takes months and multiple teams | Shippable in days |
Sum the three scores. Use cases with the highest totals go first.
Note: Effort is scored 1 for hard, 5 for easy - the way we want the weighting to run. Higher score = faster to ship = cheaper risk.
Quick wins we always ship early
These score high on business impact + low on effort, and they're usually the first things to hit production:
- Welcome email / series - High impact on retention, low effort. Most opens of any email type. Build this first.
- Transactional emails (order confirmation, shipping notification) - High impact, medium effort. Customer expects them; missing them creates support tickets.
- Re-engagement campaign for inactive subscribers - Medium impact, low effort. A simple "we miss you" send lifts 3-5% of the dead list.
- Abandoned cart email - High impact for commerce clients, medium effort if the cart data feed is available.
Things we defer to phase 2+
These need discovery + infrastructure first, or don't pay back early:
- Recommendation engine (product recommendations) - High effort, requires Einstein Recommendations or custom model + trained data. Revisit once the data model is mature.
- Real-time personalization - Requires Interaction Studio / Personalization integration + tagged website events. Multi-month build.
- Custom Preference Center - Every client asks, few clients actually need custom over the default. Ship default first; customize when the default falls short.
- Loyalty program emails - Depends on loyalty system integration. Often blocked on upstream work.
Running the exercise with the client
The actual workshop takes 45 minutes:
- List use cases - 5 minutes. Client dumps every idea onto sticky notes (digital or physical).
- Score each - 15 minutes. Walk through the 3 columns with the client. Their number, not yours - you explain the criteria, they score.
- Sort by total - 5 minutes. Stack-rank.
- Draw the line - 15 minutes. Budget permits N use cases in this phase; everything below the line goes to the backlog with its score recorded.
- Commit - 5 minutes. Get verbal buy-in on what phase 1 ships. Write it down.
The part that matters is step 4. The client resists cutting anything, and 30 minutes in someone says "but we really need X too." That's when you show them X scored a 7 versus the top-of-list scoring 14 - and the trade-off becomes concrete instead of political.
Two traps we watch for
Trap 1: Shiny-object inflation on Business Impact
Clients tend to score everything a 5 on Business Impact. Everything feels important when it's on a list. Push back by asking "if we shipped this alone and nothing else, would the quarter be a success?" That forces real 5s.
Trap 2: Under-scoring Effort
Sales engineers over-estimate Effort (it looks like work) and under-estimate Business Impact (it's not their job). Involve a client stakeholder for scoring Effort. They usually know their team's actual velocity better than the vendor.
Takeaway
Prioritization isn't about finding the objectively best use case. It's about forcing a conversation that otherwise gets deferred until the budget runs out and everyone is surprised. When the client has 20 ideas and the team is already sprinting, this framework buys everyone 3 months of sanity.
Need help prioritizing your SFMC roadmap? Our Salesforce team runs kickoff workshops and phase planning on production engagements. Get in touch ->
See our full platform services for the stack we cover.
