Why These Frameworks Matter
Most presentations fail not because the data is wrong — but because the narrative is missing. Numbers don't speak for themselves. Dropping a chart on a slide and expecting the audience to draw the right conclusion is the single most common mistake PresentHub sees across the decks we've reviewed. These five frameworks give your data the structure it needs to drive a decision — not just fill a slide.
1. The "So What?" Framework — Observation, Relevance, Action
Three sentences. Start with what you observed, say why it matters to this audience, then tell them what to do next. Most presenters stop after the observation ("our conversion rate dropped 14% this quarter") and leave the audience to figure out the rest. The "So What?" test forces you to complete the thought: "...which means we're losing roughly 800 paying customers per month...so we need to pause paid campaigns and audit landing page copy before the next board meeting."
Apply the "So What?" test to every data point before it makes it into the deck. If you can't complete the second and third sentences, the data point shouldn't be there. This is a pre-flight check, not a framework for structuring whole slides — but it's arguably the most useful filter of the five.
When it fails: When the action isn't yours to take. If you're presenting to a team that doesn't control the lever you're recommending, the "Action" step falls flat. In those cases, end with the implication, not a directive.
2. McKinsey's Pyramid Principle — Answer First, Evidence Second
Most presenters build to a conclusion: they show the data, walk through the analysis, then land on the recommendation at the end. McKinsey's Pyramid flips it — lead with the answer, then back it up. "We should exit the APAC market" goes on slide one. Slides two through five prove it. This works especially well for executive audiences who have no patience for suspense and who will interrupt with "so what's the recommendation?" before you get there anyway.
When it fails: When your answer is controversial or the decision-maker hasn't been primed. Leading with a conclusion can trigger defensiveness before you've built the case. In those situations, save the Pyramid for the written follow-up and use a narrative arc (see Framework 4) in the live room.
3. The OIA Framework — Observation, Insight, Action
OIA looks similar to "So What?" but the middle term changes everything. Where "So What?" asks "why does this matter?", OIA asks "what does this mean?" — a more analytical question that requires actual diagnostic work.
Observation: Customer retention dropped 8% in Q2. Insight: Churn is concentrated in users who didn't complete onboarding in their first 30 days — not in the broader user base. Action: Redesign the onboarding flow to hit a 90% 30-day completion rate before Q4.
The insight is the hard part. It's the step that separates analysts from advisors. Skipping it — jumping from observation directly to action — is why so many recommendations feel arbitrary or unconvincing to audiences. If your slide only has an observation and an action, you haven't done OIA; you've done guessing.
When it fails: When you genuinely don't have the insight yet. Don't fake it. Present the observation, state that you're still investigating, and commit to a date for the follow-up.
4. The Data-to-Story Arc — Context, Tension, Resolution
Borrowed from screenwriting. Context sets the baseline ("Our conversion rate held at 3.2% for three consecutive quarters"). Tension introduces the disruption ("Q3 dropped to 1.8% — our lowest since 2022"). Resolution is the path forward ("Audit reveals paid traffic quality dropped after targeting changes — reverting to Q2 parameters should recover baseline within 60 days").
This arc works well in board presentations and investor updates where you need to hold attention across 20-plus slides. It's also highly effective for post-mortems, where the tension is already known and the audience wants to understand the cause and fix.
When it fails: When you don't have a resolution. If you're still diagnosing a problem, presenting a half-formed arc feels incomplete and erodes confidence. Use OIA instead and return with the resolution at a follow-up checkpoint.
5. The 1–3–1 Framework — One Idea, Three Supports, One Takeaway
Probably the most useful framework for structuring individual slides. One headline stating the main point. Three supporting data points, visuals, or examples. One closing sentence that reinforces the takeaway.
It forces ruthless prioritization. If you can't find exactly three supports, you either don't have enough evidence or you're trying to say too many things on a single slide. In our review of AI-generated presentations across more than a dozen tools, slides that follow a 1–3–1 structure consistently receive higher clarity scores in audience feedback — the headline tells you what to think, the three supports prove it, and the takeaway tells you what to remember.
The 1–3–1 also scales to full-deck architecture for short pitches: one problem, three reasons it's urgent, one solution. It's the skeleton of a tight five-minute pitch.
When it fails: When the subject is genuinely complex and three supports aren't enough. In those cases, use a separate methodology slide rather than cramming more into the 1–3–1 structure.
Key Takeaway
These five frameworks aren't mutually exclusive — most strong presentations layer them. Use the Pyramid for your executive summary, OIA for your diagnostic slides, and 1–3–1 to structure each individual slide. Run the "So What?" test on every data point before it makes the final cut. Use the Data-to-Story Arc when you need to hold a room through a longer narrative. The biggest mistake is picking one framework and forcing everything through it. Data storytelling is situational — the right structure depends on what your audience needs to decide and how much they already know.
