What to Track in Your Quantified Self Setup (Without Overcommitting)

The single biggest reason people quit personal tracking is not laziness. It is overcommitment. They open a new app, get excited, and set up twenty fields the first evening. By week three the app feels like a chore. By week six it is uninstalled.

The smartest tracking setups are absurdly small at first. Three fields, sometimes five, almost never more. This article is about how to pick those three to five fields well, so the habit sticks and the data is actually useful when you look back.

If you have not seen our calm guide to what personal analytics actually is, start there first. This article picks up where that one stops, at the question every new tracker asks: what should I actually log?

The trap of tracking everything

Adding one more field takes ten seconds. Why not capture screen time, water, steps, mood, energy, focus, stress, three habits, a journal note, and your weight while you are at it? Because the cost is hidden, and it shows up later.

• Logging friction adds up. Thirty seconds a day across ten fields becomes five minutes by field twenty. You will skip it.
• Data quality drops with field count. Honest answers to twenty questions are rarer than honest answers to four.
• Patterns get noisier, not clearer. More fields means more pairs, more coincidences, and more work to find the signal.

A useful rule: if you would not bet money that you will still be logging this in eight weeks, do not start in week one.

”What should I track” is really three questions

When someone asks what they should track, they usually have not yet split that question into its parts. Splitting it is most of the work.

1. What do you want to understand

Tracking without a question is journaling without a prompt. It is fine, but it rarely produces the moment where the data tells you something. So before you pick a field, name the question.

Examples that produce useful setups:

• “Why is my focus so uneven from day to day?”
• “Do I actually feel different when I skip my workout, or am I imagining it?”
• “Is caffeine helping me or just delaying the slump?”
• “Are my low-energy days clustered in one part of the week, or spread out?”

Vague questions like “am I healthy?” or “am I happy?” do not narrow the field set. Specific ones do.

2. What can you measure fast and honestly

A field that takes more than thirty seconds a day will lose. So will a field that requires you to guess after the fact. Some quick filters:

• Can I log it in one tap or a few digits? If not, simplify the field.
• Will I remember it accurately at logging time? If not, the data is fiction.
• Does it have an honest unit? Hours of sleep is honest. “Sleep quality” needs to be a bounded scale, not a free-text guess, or it is going to drift.

If you cannot answer yes to all three, the field will not survive a month.

3. What pairs well with what else

This is the question most new trackers skip. The interesting patterns in personal analytics sit between fields, not inside them. A single sleep number tells you very little. A sleep number next to your focus rating the next morning is a small story.

So when you pick three to five fields, you are not really picking three to five things. You are picking the pairs you want to be able to compare. Sleep alone is less useful than sleep next to something that might be affected by it.

A rough heuristic: at least one of your fields should be a plausible cause, and at least one should be a plausible effect. With three fields you usually get two or three meaningful pairs, which is more than enough to find your first real connection.

Starter trios that work

If the open-ended choice is overwhelming, here are four small setups that cover the most common goals. Pick one, adjust to taste, and run it.

Sleep + mood + one habit

The classic. Sleep is the most common cause; mood and the habit are downstream of it. The habit can be exercise, meditation, no alcohol, screen-time cutoff, or whatever you suspect matters most. Three fields, two same-day pairs and one day-after pair to look at.

Energy + focus + caffeine

For knowledge workers. Energy and focus on a 1 to 10 scale, caffeine as a number (cups, or milligrams if you are precise). The relationship is usually less linear than you expected.

Mood + weather + activity

For understanding “off” days. Mood on a scale, weather as a categorical list (sunny, cloudy, rain, cold, hot, mixed), activity as a categorical list (deskbound, light walk, outdoor, social, rest). After a month, look at which combinations correlate with low mood.

Blood pressure + medication + activity

For anyone tracking cardiovascular numbers to share with a doctor. Not medical advice, just a clean log: a systolic and diastolic reading, a yes-or-no for medication, and a category for activity. Bring the export to your next appointment.

Pick one trio, do not blend two of them, and let the setup run for two weeks before you tweak it.

Loggr’s six field types, in plain language

Whatever you pick, the shape of the field matters. The right field type makes your data comparable later; the wrong one makes it noisy. Loggr offers six field types, and each one is good at a specific job.

• Number. Any quantity. Sleep hours, glasses of water, cups of coffee, weight, minutes meditated. Best for things with an honest unit. Up to six decimal places, so it covers everything from “8 hours” to “0.5 g of salt.”
• Scale. A bounded rating, like 1 to 10 or 0 to 5, with a chosen step size. Best for subjective ratings: mood, energy, focus, pain, stress. The bounded range keeps the data comparable week to week, which a free-text rating cannot.
• Yes / No. A single tap for “did it” or “did not do it.” Best for habits: exercised, meditated, took medication, no alcohol, screen-time cutoff respected. Cheap to log, easy to compare.
• Categorical. Pick one option from a custom list you define. Best for daily category picks: weather (sunny / cloudy / rain), workout type (cardio / strength / yoga / rest), social plans (none / casual / event). Limited options means clean stats.
• Text. A short free-form note. Best for context, not data. One sentence about your day will be priceless six months from now when you are trying to remember what made a particular week strange. Loggr can surface past entries as you type, so repeat notes are a tap.
• Blood pressure. A dedicated dual field that captures systolic and diastolic together, with its own stats and chart. Best for, well, blood pressure. Worth using because the pair belongs together.

A good starter setup uses at least two different types. The mix is what makes the patterns interesting.

What not to track

This is the section most articles skip. It is the most useful one.

Things you cannot measure honestly

If a field requires you to judge yourself after the fact, the data will drift. “How productive was I today?” is one of these. By the time you log it in the evening, the morning is fuzzy and the recent hours dominate. Replace it with something narrower: hours of focused work, number of meetings, or a 1 to 10 scale anchored to specific bands you write down.

Things you only care about as outcomes

“Happiness” by itself is not actionable. It is the thing you want; it is not the thing you can change. Track inputs instead, plus a single outcome field, and let the input fields be the ones you adjust. If you only track outcomes, your data will not tell you what is moving them.

Things a wearable already handles reliably

Loggr is a manual logging app on purpose. If your watch tracks steps and sleep more reliably than you ever will by hand, let the watch handle those. Track the things your wearable does not see: how you feel about the day, whether you actually did the thing, what category of day it was. Manual logging is a complement to passive sensors, not a replacement.

Vanity metrics

A vanity metric is a number that looks impressive but is not connected to anything you care about. They feel like progress and tell you nothing. If a field cannot affect any decision you might make, do not track it.

The two-week rule

Once you pick your three to five fields, do not add anything for at least two weeks.

This is the part that takes discipline. After three days you will think of a great extra field. After seven days you will be convinced you also need to track screen time, water, and supplements. Resist all of it.

There are three reasons.

1. You are still calibrating. A “7” for mood in week one is not the same as a “7” in week three. Adding fields before your existing ones stabilise means you are calibrating two sets at once.
2. Two weeks is the minimum for any pattern to start meaning something. Adding a new field resets the clock on its own stats and makes the comparisons less clean.
3. Most of the fields you “need to add” in the first two weeks are not actually needed. If you still want them after fourteen days of honest logging, then add them deliberately. Most do not survive that wait.

Two weeks is short. The discipline pays off in months three through twelve, when your data is genuinely worth looking at.

When to add a field later

After your two-week period, the rule is simple: only add a field when you have a specific question your current setup cannot answer.

“I want to know if my workout intensity matters, not just whether I worked out” is a specific question. Add an intensity scale or a categorical for workout type. “I should probably track water too” is not a specific question; it is a vague urge. Skip it.

When you do add a field, add one. Not three. Let it run for two more weeks before you add another. The pattern repeats: small, deliberate, calibrated.

How big should the whole setup get

A useful range for most people is between five and ten active fields. Loggr’s free plan allows five fields, which is enough to run any starter trio with room to spare. Pro lifts that limit. Even with no limit, the same principle holds: more is not better, more is just more.

If you find yourself wanting more than ten active fields, ask whether you are answering a new question with each one, or just adding out of habit. If it is the latter, prune.

A simple checklist before you log your first day

1. Write down the one question you most want to answer in the next month.
2. Pick three fields that, together, could shed light on it. Use at least two different field types.
3. Set them up with honest units and labels you will read in six months without confusion.
4. Pick a daily logging time per field. Sleep is best logged in the morning; mood is best logged at night.
5. Decide that you will not add anything new for two weeks. Write the date you are allowed to revisit.
6. Log every day for the first fourteen days, then look at your weekly summary.

FAQ

Can I track more than five things in Loggr?

The free plan allows up to five fields, total and enabled. Pro lifts that limit if you want more. For most people, five to ten active fields is the sweet spot. Beyond ten, you usually trade data quality for data quantity.

Should I copy someone else’s tracking setup?

Not exactly. You can borrow a starter trio as a starting point, but the fields should reflect your questions and your life. Copying a famous setup wholesale is a recipe for tracking things you do not care about, which is the fastest way to quit.

What if I miss a day?

Nothing breaks. Coverage drops, and that field’s stats for the week will be based on fewer days. If you remember the values, you can log a past date later. If not, leave the gap. Honest gaps are better than guessed numbers.

Do I need to log at the same time every day?

Same time per field is more important than same time across fields. Logging sleep at 8am every day is consistent, even if you log mood at 10pm. The reason: you want each field’s distribution to be measured under similar conditions, which is a per-field property.

How long before patterns become trustworthy?

Plan on a month. Weekly views are readable sooner, but a real connection between two fields usually needs at least twenty samples and enough variation to compare. The day-after relationships in particular need a few weeks of consistent logging before they settle down.

Key takeaways

• The biggest reason people quit tracking is overcommitment, not laziness. Start small on purpose.
• “What to track” is really three questions: what do you want to understand, what can you measure fast and honestly, and what pairs well with what else.
• A good starter setup is three to five fields covering at least two different field types.
• Loggr’s six field types are number, scale, yes or no, categorical, text, and blood pressure. The right shape per field makes the data comparable later.
• Skip fields you cannot measure honestly, fields that are outcomes-only, fields a wearable already handles, and vanity metrics.
• Do not add anything new for two weeks after your starter set. Only add later when you have a specific question your current setup cannot answer.

Open Loggr and add three fields right now

The shortest path is the boring one. Open Loggr and create three fields: sleep hours, mood on a 1 to 10 scale, and one habit you think matters. That is it. Run that setup for two weeks. Do not add anything else, do not switch the scale range, do not optimise the icons. Just log, every day, at a time you can keep up with.

When the two weeks are up, glance at the weekly stats and see whether any pair tells you something you would not have guessed. If it does, you have your first real personal analytics result. If it does not, you have learned that those three were not the right three, which is also a result. Either way, you will know what to do next.