From Spreadsheets to Split Times: A 6-Week DIY Data Plan for Runners

If you’ve ever stared at a Garmin export or a Strava activity file and thought, “I know this contains answers, but I don’t know how to find them,” this guide is for you. The goal of this six-week training plan is not to turn runners into software engineers; it’s to build practical data literacy so you can make better training decisions from the data you already collect. By the end, you’ll know how to export your running files, clean them in pandas, visualize trends, and turn that information into an actual improvement plan you can follow. If you want the mindset side of big goals, our guide to fitness mindset for life transitions is a helpful companion read.

This is a true DIY analytics project built for runners, not analysts. You’ll use free workshops, open-source tools, and a simple weekly structure to go from raw files to useful split times, pacing charts, and recovery insights. The best part is that this approach scales: whether you’re training for a 5K, half marathon, triathlon, or just trying to stop bonking on long runs, the same workflow can help. Think of it like the difference between owning a watch and actually understanding your watch data: one records, the other informs. And if you’re upgrading your kit too, our piece on spotting quality in an athletic jacket shows how to evaluate gear without overspending.

Why runners should learn data literacy now

Training data is only useful if you can interpret it

Most runners already have more data than they can comfortably use. Garmin, Strava, HR straps, power meters, sleep tracking, and shoe logs can create a rich picture of your training load, but only if you can turn those numbers into decisions. Without a basic analytics workflow, it’s easy to chase one-off bad splits, overreact to fatigue, or assume every plateau is a fitness problem. Learning how to analyze your own data helps you distinguish normal variation from meaningful trends. For a broader look at how structured learning opportunities can accelerate skill-building, see free data analytics workshops.

Why this matters for endurance performance

Endurance training rewards consistency more than hero sessions. When your data is organized well, you can spot whether your easy pace is drifting faster at the same heart rate, whether your long-run fade is getting smaller, or whether your cadence changes when fatigue creeps in. These are actionable signals, not vanity metrics. They help you decide when to progress volume, when to back off, and when to adjust fueling or sleep before the next key workout. Runners who use data well often train more confidently because they can see the story behind the numbers rather than guessing.

What this plan will teach you

This six-week programme focuses on the essential steps of Garmin export, Strava data handling, cleaning with pandas, visualizing with open-source tools, and turning patterns into decisions. You will not need a paid analytics platform to get started. You will need curiosity, a spreadsheet mindset, and a willingness to work through a few awkward file formats at the beginning. The result is a repeatable system you can use every training block, race build, or off-season reset. If you like the “learn by doing” format, you may also appreciate AI-enhanced microlearning for busy routines.

What you need before Week 1

Your free toolkit

You don’t need expensive software to start. A typical setup includes Python, Jupyter notebooks, pandas, matplotlib or seaborn, and optionally Plotly or Streamlit for dashboards. For file management, a folder structure and a spreadsheet are enough. If you’re interested in setup philosophy and tradeoffs, the same principles apply in tech-heavy domains like SDK selection or automation versus manual work: start with the simplest tool that can do the job reliably. In running, that means choosing tools you can maintain after the novelty wears off.

What data to collect

At minimum, track date, workout type, duration, distance, average pace, lap splits, heart rate, elevation gain, and perceived effort. If available, add cadence, temperature, power, and sleep score. The more complete the dataset, the better your comparisons, but don’t let perfection become a barrier. Your first objective is a clean, consistent dataset that covers at least six weeks of training and is detailed enough to reveal trends. If you’re also thinking about equipment purchases, our guide on value-focused purchase analysis demonstrates the same “data before decision” mindset.

How to think about the project

Treat this like a training block for your analytical skills. Each week has one main objective: export, clean, organize, visualize, interpret, and act. That structure matters because the point is not simply to create charts; it is to create feedback loops. Runners often train with progression but analyze with randomness, which is why results feel mysterious. A good analytics workflow makes your training decisions feel as deliberate as your interval sessions.

The 6-week DIY analytics programme

Week 1: Export and inventory your training data

The first week is about getting control of your raw files. Export your activities from Garmin Connect and Strava, and decide which source will be your “master record” for each workout. Garmin is often richer for physiological metrics, while Strava may be easier for social context and activity history. Keep both if possible, but don’t merge blindly until you understand the column names, formats, and missing values. This stage is similar to building any well-organized system: start with collection, then standardization. If you enjoy the workflow perspective, event-driven workflow design offers a useful analogy for how input, processing, and output fit together.

During this week, create a simple folder structure: raw_exports, cleaned_data, charts, and notes. Save every export with a date and source in the file name. Then make a tiny inventory sheet listing file type, date range, source, and completeness. This alone will save you hours later. Think of it as the runner’s version of labeling gear and nutrition so you don’t mix up recovery drink with pre-run fuel.

Week 2: Clean your data in pandas

In week two, open the files in pandas and inspect them before doing anything else. Check for column names, data types, missing values, duplicate activities, and timezone issues. This is where beginners often discover that what looked like “one run” in the app appears as two records in an export or that pace fields are stored as strings rather than numbers. Cleaning is not glamorous, but it is the difference between trustworthy analysis and misleading conclusions. The same logic applies to any data workflow, including the standards discussed in legal lessons for data use: know what you have, know where it came from, and know how it’s being transformed.

Here’s the practical goal for this week: make one tidy table where each row equals one workout and each column equals one variable. Convert times to consistent units, standardize workout labels, and create derived columns such as pace per kilometer, training load category, or intensity zone. A clean dataset should let you answer basic questions without manual clicking. Once you can filter by workout type, pace range, and effort, you’ve already unlocked most of the value.

Week 3: Build your first training dashboards

Week three is when the project starts to feel rewarding. Use matplotlib, seaborn, or Plotly to visualize weekly mileage, average pace, heart rate, and recovery markers. At minimum, create a timeline chart, a pace distribution plot, a rolling weekly volume chart, and a simple scatterplot of pace versus heart rate. The objective is not artistic perfection; it is clarity. A dashboard should answer questions quickly, just like the best voice-enabled analytics tools try to make insights easier to surface.

Focus on comparisons that matter to performance. Are your easy runs getting easier at the same pace? Are your interval paces becoming more repeatable? Do long runs show a late-session slowdown? Good dashboards reduce uncertainty by showing patterns over time. If you’ve ever admired a well-designed consumer display, the same principle of clear hierarchy appears in product thinking too, such as in product design storytelling.

Week 4: Analyze split times and workout structure

Now move from big-picture load to the micro level of split times. This is where a runner learns whether a session was actually executed as planned. Compare interval splits, threshold segments, negative splits, and pace decay in the final reps of a workout. You can calculate split consistency by measuring the spread between fastest and slowest reps or by looking at coefficient of variation. The point is to understand whether you’re pacing evenly, fading, or improving session control.

A useful exercise is to compare your best workouts from the last six weeks against your average sessions. Did you start too hard? Did heart rate drift upward despite constant pace? Did recovery runs become genuinely lighter over time? This level of analysis is similar to how elite recruiting workflows evaluate talent patterns, not just one standout result. For a striking parallel outside running, see elite data workflows in scouting, where consistency often predicts future success better than a single flashy performance.

Week 5: Turn insights into a training plan

By week five, your data should start telling a story. Now convert that story into a decision-making framework for the next six to eight weeks. If your easy pace is improving at the same heart rate, you may be ready for a small intensity progression. If your pace is holding but perceived effort is rising, you may need more recovery. If your long-run split fade is persistent, you might need better fueling or a more conservative start. This is where analytics becomes coaching rather than reporting.

Use your charts to define three rules for your next block. For example: increase weekly mileage by no more than 5-8% if recovery markers stay stable; keep one full rest day after hard interval sessions; and review long-run hydration if pace drops more than a chosen threshold in the final third of the workout. Training should be progressive but not reckless. If you’re rebuilding after burnout or a life change, our guide on goal resilience can help frame the process.

Week 6: Package your results into a repeatable system

The final week is about making the workflow sustainable. Document the steps you used, save your notebook, and create a checklist for future training blocks. Label where the raw exports come from, what cleaning steps are always needed, and which charts you check every Sunday. If you make the process repeatable, it becomes a habit instead of a project. That matters because the best analytics system is the one you’ll actually keep using.

In practical terms, your “finished” system should include one file for raw data, one notebook for cleaning, one dashboard view for weekly review, and one note template for training decisions. The outcome is a personal performance cockpit you can revisit after every key cycle. When runners get this right, they stop asking “Was I fit?” and start asking “What pattern emerged, and what should I do next?”

How to turn raw exports into useful runner metrics

Best metrics for everyday endurance training

Not every metric deserves equal attention. For most runners, the highest-value measures are weekly distance, session count, pace at a given heart rate, long-run consistency, tempo execution, and recovery trends. Those data points tell you more about adaptation than a dozen vanity charts. Consider tracking cadence only if it influences injury risk, efficiency, or run form. Focus on measures that support decisions.

Heart rate drift is especially useful for endurance athletes because it can reveal fatigue, heat stress, insufficient fueling, or improved aerobic efficiency. Split consistency in intervals can tell you whether your pacing skills are improving. Weekly load trends help prevent sudden jumps that outpace adaptation. If you want to compare training efficiency against a broader standard of “what is worth keeping,” the logic resembles product evaluation in buy-versus-keep decisions: the right choice depends on value, not hype.

How to avoid misleading conclusions

One bad workout does not define fitness, and one great workout does not guarantee readiness. Context matters. Heat, terrain, sleep, travel, hydration, and race pressure all affect pace and heart rate. That’s why the best runners compare like with like: similar routes, similar conditions, and similar effort levels. Without that discipline, your dashboard may look impressive while telling the wrong story. Careful comparison is the foundation of trustworthiness in any analysis workflow, from market analysis to training review.

How to define personal baselines

Your baseline should be based on your own recent history, not someone else’s training screenshots. Pick a four-to-six-week window and calculate your normal ranges for easy pace, average heart rate, weekly volume, and recovery time after hard sessions. Once you know your range, small changes become visible. If your average easy-run heart rate drops at the same pace, that is meaningful. If your threshold interval split standard deviation shrinks, that’s also meaningful. Baselines help you separate noise from adaptation.

Pro Tip: The most useful dashboard for most runners is not the fanciest one. It’s the one that shows weekly load, easy-run heart rate, long-run split fade, and one recovery marker in under 30 seconds.

Free workshops and open-source learning resources that actually help

What to look for in a workshop

Choose workshops that teach hands-on Python basics, pandas workflows, and data visualization with real datasets. You want practical work, not just slides. The best free sessions usually offer notebooks, downloadable code examples, and enough structure for beginners to follow without friction. If you’re comparing learning formats, our source-grounding article on free workshops in data analytics is a good starting point for understanding what effective beginner-friendly training looks like.

What open-source tools to prioritize

Start with pandas for cleaning and transformation, matplotlib or seaborn for static charts, and Plotly or Streamlit when you want interactive dashboards. Add Jupyter notebooks so you can document each step and preserve your thinking. Later, if you want automation, you can schedule exports and refresh charts, but don’t begin there. Simplicity first, automation second. This principle also shows up in other domains where setup complexity can overwhelm results, like low-stress automation systems or operational pipelines.

How runners can learn faster than they think

Runners are already used to progressive overload, feedback, and repetition. That makes data literacy more approachable than it may seem. One session can teach you how to load CSVs, another can teach aggregation, and another can show visualization principles. If you prefer to learn in short bursts, use microlearning habits: 30 minutes to export, 30 minutes to clean, 30 minutes to chart. Repeat weekly until the workflow becomes familiar. This kind of structured learning mirrors the real-world logic behind microlearning design.

Detailed comparison: tools, use cases, and tradeoffs

The table below shows how runners can think about their tool choices. Use it as a practical decision aid rather than a perfect ranking. The best stack is the one you can keep up with consistently, not the one with the most features.

Case study: how a runner might use this system

Scenario: the plateaued 10K runner

Imagine a runner who has been stuck at the same 10K time for nine months. Their mileage is consistent, but the workouts feel harder than they used to. After exporting six weeks of Garmin data, they notice that easy runs at the same pace now sit several beats higher in heart rate than they did earlier in the block. They also find that their final reps in interval sessions are drifting slower, even when the first reps look fine. That pattern suggests accumulated fatigue or insufficient recovery rather than a total lack of fitness.

What the data suggests

The runner reviews weekly volume and sees a small increase in intensity density: too many moderate efforts, not enough true recovery. They also notice that long runs are often done after poor sleep or under-fueled mornings. With that insight, they adjust the plan: one fewer quality day per week, more deliberate easy days, and a fueling strategy for runs over 75 minutes. Three weeks later, the easy pace at the same heart rate improves and interval split variation narrows. This is the practical power of DIY analytics: it changes behavior, not just understanding.

Why this kind of story matters

Good training analytics should make you more confident, not more anxious. The point is to identify patterns that lead to small, sustainable improvements. You are not trying to turn every run into a laboratory experiment. You are trying to create a durable feedback system that respects your body and your schedule. If you’re building resilience for demanding goals, our discussion of goal-focused fitness mindset reinforces why patience matters.

Common mistakes runners make with data

Tracking too much too soon

Many runners start with an overly ambitious dashboard and quickly burn out. If you try to track every metric available, you’ll spend more time organizing than training. Begin with the variables that directly affect your decisions. Add extra layers only after you have a stable routine. This same “start narrow, then expand” principle is how many value-conscious buyers avoid overpaying for features they do not need, as seen in guides like subscription alternatives.

Confusing correlation with causation

Just because a hard session followed a bad night of sleep does not mean sleep alone caused the poor workout. Maybe the route was hillier, the weather was warmer, or you were already carrying fatigue from the prior week. Use data to generate hypotheses, then test them over time. Keep notes on context and subjective effort. That combination of numbers plus narrative is what makes the analysis trustworthy.

Ignoring recovery data

Training adaptations happen during recovery, not just during workouts. If you ignore sleep, resting heart rate, soreness, and perceived fatigue, your dashboard will miss the conditions that make performance possible. Recovery signals often tell you whether to push, maintain, or back off. In endurance sport, that can be the difference between building fitness and digging a hole. If you’re curious about wellness systems that integrate data and behavior, see technology-forward wellness centers.

How to use your findings for the next training block

Create a simple decision tree

After six weeks, build a decision tree that links data patterns to training actions. For example: if weekly mileage is stable and recovery is good, increase volume slightly; if easy pace is slowing at the same heart rate, reduce intensity; if interval splits are inconsistent, shorten reps or lengthen recovery. Keep the rules few and clear. A training plan becomes more effective when it gives you decisions, not just inspiration.

Set one main focus for the next block

Do not try to fix everything at once. If your data shows poor long-run pacing, make that the emphasis. If it shows heavy fatigue, prioritize recovery and load management. If it shows strong aerobic improvement but weak top-end speed, shift the focus to speed endurance. One focus per block is enough to move the needle while keeping the plan manageable. For runners balancing performance with everyday life, the broader productivity lens in pilot planning is surprisingly relevant.

Build accountability into the process

Share your dashboard with a training partner, coach, or local running group if that helps you stay consistent. Accountability is easier when the metrics are simple and the goals are clear. A weekly review ritual works better than sporadic deep dives. The goal is not to become obsessed with charts; it is to create enough structure that good decisions happen by default. If you appreciate systems thinking, you may also like structured transformation roadmaps and how they sequence change.

FAQ

Final takeaway: turn training into a feedback loop

The smartest training plan is not the one with the most complicated weekly structure; it’s the one that helps you make better decisions repeatedly. By learning how to export your Garmin and Strava data, clean it with pandas, visualize key patterns, and act on what you see, you create your own endurance feedback loop. That loop can improve pacing, reduce wasted effort, and make your progression more intentional. It can also make training feel less mysterious and more rewarding.

If you want to keep building that system, revisit your data after each training block, compare the new patterns to the last one, and adjust only one or two variables at a time. That’s how durable improvement happens. And when you’re ready to expand the process, go back to the broader learning resources in free data analytics workshops and the workflow thinking behind operational pipelines. The athlete who understands their own data becomes harder to derail, easier to coach, and much more likely to keep improving.

Related Reading

• How to Spot Quality in an Athletic Jacket Without Paying Premium Prices - Learn what actually matters when choosing durable training gear.
• FIT TO SELL: How Fitness Mindset Helps You Navigate Life Transitions (and Big Goals) - Build the mindset that supports long-term progress.
• Lifelong Learning at Work: Designing AI-Enhanced Microlearning for Busy Teams - Borrow a microlearning system for your analytics habits.
• Designing Event-Driven Workflows with Team Connectors - See how structured workflows improve consistency.
• Estimating ROI for a Video Coaching Rollout: A 90-Day Pilot Plan - Useful for thinking about training interventions like pilots.