Building an R01 Approach Section Reviewers Can Follow: Methods, Milestones, and Contingency Plans

Of all the scored criteria on an NIH R01 application, Approach has historically carried the most weight in determining overall impact scores. That relationship has only sharpened under the Simplified Peer Review Framework that took effect for applications due on or after January 25, 2025, which folds Approach into its own dedicated factor -- Factor 2: Rigor and Feasibility -- scored on the familiar 1-to-9 scale. Translation: the twelve pages you devote to Research Strategy will make or break your application, and the Approach section is where most of those pages live.

Yet the Approach is also where the most R01 applications hemorrhage points. Study sections routinely cite the same failures: vague methods, missing power analyses, no contingency plans, preliminary data that fails to establish feasibility. These are not exotic weaknesses. They are structural problems -- the result of applicants who know their science cold but have never been taught how reviewers actually read and score twelve pages of dense methodology.

With R01 success rates hovering around 22 percent across NIH and many institutes running interim paylines in the single-digit percentiles (NIAID's FY 2025 interim payline for established investigators sits at the 8th percentile), the margin between a fundable score and a "revise and resubmit" often comes down to how clearly the Approach section communicates that the work can actually be done.

What "Rigor and Feasibility" Means Under the New Framework

Before January 2025, reviewers scored five separate criteria: Significance, Investigators, Innovation, Approach, and Environment. The new Simplified Peer Review Framework reorganizes these into three factors. Factor 1 (Importance of the Research) absorbs Significance and Innovation. Factor 3 (Expertise and Resources) covers Investigators and Environment -- but as a binary sufficiency check, not a numerical score. Factor 2 (Rigor and Feasibility) stands alone, scored 1 to 9, and maps directly onto what used to be the Approach criterion.

The shift matters for grant writers because Factor 2 now asks reviewers to evaluate two things simultaneously: the likelihood that the proposed studies will produce compelling, reproducible findings (rigor), and whether the studies can be executed well within the proposed timeframe (feasibility). That dual mandate means your Approach section needs to do more than describe experiments. It needs to convince reviewers that every experiment is designed to minimize bias, powered to detect meaningful effects, and backed by a realistic plan for when things go sideways.

The old system allowed a strong investigator score or environment score to partially compensate for a middling approach score in the overall impact assessment. Factor 3's binary format makes that harder. If your team and resources are deemed sufficient, they earn a check mark -- not a 2 or 3 that might pull your overall impact up. The Approach section now stands more exposed.

Organizing Twelve Pages So Reviewers Do Not Get Lost

The Research Strategy page limit for an R01 is 12 pages, covering Significance, Innovation, and Approach (including preliminary data or a progress report for renewals). Images, figures, graphs, and tables all count against this limit. Most successful applications allocate roughly one to two pages for Significance, one page for Innovation, and the remaining eight to ten pages for Approach plus preliminary data.

Within the Approach section itself, the organizational principle that earns the best scores is simple: mirror the structure of your Specific Aims page. Each aim gets its own subsection. Within each aim, follow a consistent internal structure:

Rationale. A brief paragraph stating why this aim is necessary and how it connects to the overall hypothesis. Reviewers read dozens of applications per study section. Restating the logic at the top of each aim prevents them from having to flip back to your Specific Aims page.

Methods. The core of each aim subsection. Describe the experimental design, model systems, reagents, measurement techniques, and analytical approaches with enough specificity that a reviewer in your field could evaluate whether they would work. Vagueness here is the single most common weakness cited in summary statements.

Expected results. State what you anticipate finding and why. This is not padding -- it demonstrates that your experiments are hypothesis-driven rather than exploratory fishing expeditions.

Potential pitfalls and alternative approaches. Address what could go wrong and what you would do instead. More on this below.

Milestones and timeline. Concrete deliverables tied to specific time points in the project period.

Bold headers, consistent numbering, and white space are not cosmetic choices -- they are navigational tools for reviewers who may be reading your application at 11 PM the night before study section. NIH explicitly recommends using bold headers or an outline/numbering system applied consistently throughout.

Preliminary Data: Enough to Prove Feasibility, Not Enough to Prove the Hypothesis

The question every R01 applicant asks -- how much preliminary data is enough? -- has no universal answer, but the framework for thinking about it is consistent. Preliminary data serves one purpose in the Approach section: demonstrating that your proposed methods are feasible and that your model system behaves as expected. It is not there to prove your hypothesis. If you already had proof, you would not need the grant.

NIAID's guidance frames it well: you want enough preliminary evidence to inform each of your aims, making clear that there is a strong rationale for each aim and that you know what you are actually trying to accomplish. For a three-aim R01, that typically means at least one figure or dataset per aim showing that the foundational method works in your hands, in your model, under conditions relevant to the proposed experiments.

The calibration changes with risk. The more paradigm-shifting your hypothesis, the more preliminary data you need. A proposal using well-established assays to test an incremental hypothesis can get away with a page of supporting data. A proposal introducing a novel imaging technique to study a poorly understood pathway needs to demonstrate that the technique produces reliable, interpretable results -- ideally with validation against a gold standard.

Where applicants frequently go wrong is front-loading all their preliminary data before the Approach section, creating a wall of figures that reviewers must absorb before encountering any methods. A more effective strategy weaves preliminary data into each aim's subsection, placing it immediately before the methods it supports. This way, a reviewer reads the rationale, sees the preliminary figure demonstrating feasibility, and then reads the proposed experiment -- a narrative arc that builds confidence aim by aim.

One tactical note: figures consume page real estate fast. A single multi-panel figure with a detailed legend can eat half a page. Design your preliminary data figures to be compact and high-information-density. Combine related panels. Use figure legends that do double duty, explaining both what the data show and how they support feasibility of the proposed work.

Power Analysis and Statistical Planning That Reviewers Actually Believe

Statistical rigor is not optional in the Approach section, and reviewers have grown increasingly sophisticated about spotting power analyses that exist only to justify a convenient sample size. The standard expectation is 80 percent statistical power to detect a clinically or scientifically meaningful effect size -- but the key word is "meaningful." Reviewers want to see that the effect size you powered for is grounded in prior data or published literature, not reverse-engineered from the number of animals you can afford.

A credible power analysis includes four components: the statistical test you plan to use, the effect size you aim to detect (with justification), the alpha level (typically 0.05), and the resulting sample size. If your preliminary data or published literature support the chosen effect size, cite it explicitly. If no prior data exist -- common in truly novel research -- acknowledge that and explain how you derived your estimate, whether from related assays, logical inference, or a pilot study you plan to conduct in the early months of the grant.

Common mistakes that draw reviewer criticism:

Overly optimistic effect sizes. If published studies in your area show effect sizes of 0.3 to 0.5, and you power your study for an effect size of 1.0, reviewers will question your assumptions. Smaller effect sizes require larger sample sizes, and pretending otherwise does not make the problem go away.

Missing attrition calculations. For longitudinal studies or clinical trials, your power analysis must account for expected dropout rates. A study powered for 100 participants that expects 20 percent attrition needs to enroll 125.

No plan for multiple comparisons. If your study involves multiple primary endpoints or multiple group comparisons, reviewers expect to see how you will control the family-wise error rate -- Bonferroni correction, false discovery rate, or another appropriate method.

Generic statistical language. "Data will be analyzed using appropriate statistical tests" is a red flag. Name the tests. Justify them. If you plan to use mixed-effects models for repeated measures, say so and explain why that approach suits your data structure.

For clinical and translational research, the statistical plan should also address interim analyses, stopping rules (if applicable), and how you will handle missing data. These details signal methodological maturity and give reviewers confidence that you have thought beyond the idealized scenario.

Writing Contingency Plans That Do More Than Check a Box

NIH review criteria explicitly ask reviewers to evaluate whether you have presented "potential problems, alternative strategies, and benchmarks for success." Despite this, contingency planning is routinely identified as the weakest component of R01 Approach sections. Many applicants treat it as an afterthought -- a sentence or two at the end of each aim acknowledging that "if this does not work, we will try something else."

Effective contingency plans do three things. First, they identify specific, plausible failure modes -- not generic risks, but the particular ways an experiment might fail given its design and the biology involved. Second, they describe concrete alternative approaches that are genuinely different from the primary method, not minor parameter adjustments. Third, they include decision criteria: how will you know when to pivot?

Consider structuring contingency plans as branching logic. If Aim 1 experiments yield result X (the expected outcome), proceed to Aim 2 as planned. If Aim 1 yields result Y (a plausible alternative), pivot to modified Aim 2 with adjusted parameters. If Aim 1 produces no interpretable signal, implement backup method Z, which you have already validated in a different context (cite your preliminary data).

Flowcharts and decision trees are underused tools in R01 applications. A well-designed decision tree can communicate in a quarter-page what would take a full page of prose, and it signals to reviewers that you have genuinely mapped the experimental landscape rather than narrating a best-case scenario.

The best contingency plans also address what happens between aims, not just within them. If Aim 1 is a methodological development aim and Aim 2 depends on its success, reviewers need to know your plan for Aim 2 if the Aim 1 method falls short of expectations but still produces usable data. Partial success scenarios are more common than complete failure, and addressing them shows the kind of experience that comes from actually having run complex multi-year projects.

For early-stage investigators, contingency plans carry extra weight. Without a long track record, you need to demonstrate methodological sophistication through your planning. Show that you understand what can go wrong -- it is one of the most direct ways to prove you are ready to lead an independent research program.

Milestones and Timelines That Prove You Can Finish in Five Years

A timeline is not a decoration. It is a feasibility argument compressed into a visual format. Under the new Simplified Framework, reviewers scoring Factor 2 are explicitly asked whether the studies can be done well within the proposed timeframe. A vague or missing timeline invites a feasibility critique that directly lowers your Rigor and Feasibility score.

The most effective timelines take the form of a Gantt chart or table showing each aim's major activities mapped against the five-year project period (or four years, or three, depending on the proposed duration). Within each aim, break activities into phases: method optimization, data collection, data analysis, manuscript preparation. Show where aims overlap -- parallel execution demonstrates efficiency and argues for the full budget request.

Milestones should be concrete and measurable. "Complete data collection for Aim 2" is a milestone. "Make progress on Aim 2" is not. Good milestones function as go/no-go decision points: if you have not reached a particular milestone by the end of Year 2, what does that mean for the rest of the project? Tying milestones to your contingency plans strengthens both sections.

A few specific practices that improve timeline credibility:

Account for startup time. Hiring a postdoc, breeding a mouse colony, obtaining IRB approval for a clinical protocol -- these activities take months. If your timeline shows data collection starting in Month 1, reviewers who have run labs will notice.

Show the critical path. If Aim 3 cannot begin until Aim 1 produces a validated reagent, make that dependency explicit. Reviewers respect honesty about dependencies more than they respect an artificially compressed timeline.

Budget alignment. Your timeline should be consistent with your budget. If you request a postdoc for all five years but your timeline shows their primary work concentrated in Years 2 through 4, a reviewer may question the Year 1 and Year 5 justification.

Rigor and Reproducibility: The Elements Reviewers Now Expect to See

Since 2016, NIH has required applicants to address four elements of rigor and reproducibility, and under the new framework these feed directly into the Factor 2 score. They are not box-checking exercises -- they are substantive components of your Approach section.

Scientific premise. Briefly evaluate the rigor of the published work that forms the foundation for your proposed research. If key prior studies have limitations (small sample sizes, lack of replication, potential confounders), acknowledge them and explain how your proposed work accounts for or overcomes those limitations.

Rigorous experimental design. Describe how you will minimize bias -- blinding, randomization, inclusion/exclusion criteria, positive and negative controls. For animal studies, address the sex of animals as a biological variable (required since 2016) and justify any single-sex designs.

Consideration of biological variables. Beyond sex, address other relevant biological variables: age, strain, cell passage number, time of day for behavioral assays, seasonal variation for field studies. The specific variables depend on your research area, but reviewers expect you to have thought about which variables matter and how you will control or account for them.

Authentication of key biological resources. If your proposed experiments depend on specific cell lines, antibodies, transgenic animals, or other biological resources, describe how you will verify their identity and quality. Mis-identified cell lines and non-specific antibodies have been well-documented sources of irreproducibility.

These elements should be woven into each aim's methods description, not siloed in a separate "Rigor" paragraph. Describing your randomization scheme in the context of the specific experiment it applies to is more convincing than a generic statement about your commitment to rigorous design.

Twelve Pages, One Argument

The most fundable R01 Approach sections share a quality that transcends any single technique or trick: they read as a single, coherent argument. The Specific Aims page states the hypothesis. The Significance section explains why it matters. The Innovation section explains what is new. And the Approach section -- aim by aim, method by method, contingency by contingency -- builds the case that this particular team, with this particular design, will produce rigorous answers to the questions posed.

Every element discussed here -- preliminary data, power analyses, milestones, contingency plans, rigor considerations -- serves that argument. When they are present and well-executed, reviewers can follow the logic from hypothesis to expected outcome without stumbling over gaps. When any one is missing or perfunctory, the argument develops a crack, and cracks in feasibility arguments tend to widen during study section discussion.

The investigators who consistently earn fundable scores are not necessarily the ones with the most publications or the largest labs. They are the ones who write Approach sections that answer the reviewer's core question before it is asked: can this work actually be done, and will it produce results we can trust?

Granted helps research teams assemble stronger proposals by organizing methods, milestones, and compliance requirements so the science stays front and center.