What Are Stems? (Music, Audio & Video Explained)

If you’ve ever worked with a mix engineer, received a remix pack, or delivered audio for a sync deal, you’ve probably encountered the term “stems.” But what exactly are they, and why do they matter so much in modern music production?

Whether you’re a producer sending files to a mastering engineer, a video editor scoring a YouTube video, or an artist preparing tracks for live playback, understanding stems will save you time, prevent costly miscommunications, and give you more control over your audio.

In this guide, you’ll learn everything you need to know about stems in music—from the core definition to practical workflows for creating, organizing, and delivering them in 2026 and beyond.

Introduction to Music Stems

Music stems are the backbone of modern music production, giving artists and producers the ability to break down a song into its core components—such as drums, bass, vocals, and other instruments. By separating a track into these individual elements, music stems provide unparalleled creative control over every aspect of a song. This means you can tweak the drums without affecting the vocals, or adjust the bass without altering the rest of the mix.

In practical terms, stems are groups of related tracks that, when combined, recreate the full song. For example, a drum stem might include all the drum sounds, while a vocal stem contains all the vocal parts. This approach allows for more flexible mixing, easier collaboration, and the ability to create alternate versions or remixes with ease.

For artists and producers, working with music stems opens up new possibilities in music production. You can experiment with different arrangements, isolate individual elements for creative effects, or share stems with collaborators for remote mixing and mastering. Whether you’re fine-tuning a mix or preparing your music for release, understanding and utilizing stems gives you more control over your sound and helps you achieve professional results.

Quick answer: what are stems in music?

Stems are grouped, mixed audio files that represent sections of a song—like a drum stem, vocal stem, or bass stem—that together recreate the final mix. Think of them as mini-mixes: each stem contains multiple tracks that have been processed, balanced, and bounced down into a single stereo file.

Here’s the critical distinction many artists miss: stems are NOT the same as raw multitracks. When you export stems, you’re delivering processed audio that already includes your EQ, compression, panning, and automation. Multitracks, on the other hand, are the individual elements straight from your recording studio—every mic, every take, every layer.

For example, a 2025 pop song might have 120+ individual tracks in the full DAW session—including separate mics for kick, snare, hi-hats, toms, overheads, room mics, multiple vocal takes, guitar layers, and synth patches. But when that same song is delivered as stems, it might be just 6 files: drums, bass, guitars, synths, lead vocal, and background vocals.

Stems are used daily across nearly every corner of audio production:

• EDM producers receiving remix packs from labels
• Film and TV composers delivering music for dialogue editing
• Podcasters separating dialogue from music beds
• Worship teams running backing tracks during live playback
• Many artists preparing alternate versions for streaming platforms

Quick recap for skimmers:

• Stems = grouped, processed audio files (usually stereo WAV)
• NOT the same as raw multitracks or individual tracks
• When combined, stems recreate the entire mix
• Used for remixes, mastering, sync, live shows, and more

What are stems? (Core definition & basics)

Stems are mixed groups of related tracks—all drum mics, all backing vocals, all guitars—bounced into a single file. These component recordings collectively form the entire track, making stems essential in music production for mixing, mastering, and tailoring music to visual content. They’re typically delivered as stereo WAV files at 24-bit with sample rates of 44.1kHz, 48kHz, or higher, depending on the project requirements.

The defining characteristic of stems is this: when you line up all stems from bar 1, beat 1, they should sum to recreate the final mix (or come very close, assuming no heavy mix bus processing creates phase differences).

In contemporary production, you’ll typically encounter these stem groups:

• Drums (kick, snare, toms, overheads, room mics combined)
• Percussion (shakers, tambourines, claps, auxiliary hits)
• Bass (bass guitar, synth bass, sub layers)
• Guitars (rhythm, lead, acoustic, all DI and amp tracks)
• Keys/Synths (pianos, pads, leads, arps)
• Lead vocal (main vocal with processing)
• Backing vocals (harmonies, doubles, ad-libs)
• FX/Ear candy (risers, impacts, transitions, sound effects)

These processed groups are often referred to as song stems, which allow producers, DJs, and content creators to access and creatively use different parts of a song for remixes and video production.

Let’s make this concrete. Imagine a 2023 pop track with 90+ tracks in the original session—including an entire drum kit recorded with 12 mics, stacked vocal harmonies, layered synths, and multiple guitar parts. When the producer delivers this to a label or mastering engineer, they might reduce it to just 8 stems that capture the complete sound of the whole track.

Unlike untouched recordings, stems usually include all the processing that was applied during mixing. Your drum stem contains the compression, EQ, saturation, and parallel processing you used. Your vocal stem includes the vocal reverb, delay throws, and de-essing. This is what makes stems so useful—they preserve your creative decisions while still offering flexibility.

Engineers may also create their own stems by separating individual instrument tracks or effects, and different engineers have preferences for how they group or process these stems, such as printing them wet or dry, or grouping by instrument type.

Stems vs. tracks vs. multitracks

Three terms get thrown around constantly in audio production: track, multitrack, and stem. Understanding the differences will prevent confusion when you’re receiving stems from collaborators or sending stems to clients.

Track: A single audio element in your session. This could be a snare top mic, a lead vocal take, a guitar DI recording, or a synth patch. Tracks are usually mono or stereo and represent one discrete audio source.

Multitracks: The complete collection of individual tracks recorded for a song. This includes every drum mic, every guitar layer, every vocal double, and every synthesizer part. Multitracks are generally unprocessed or only lightly processed—they’re the raw material a mix engineer needs to build a mix from scratch.

Stems: Grouped, processed submixes created after mixing. All drums combined into one drum stem. All guitars into one guitar stem. Stems carry the processing decisions you made during mixing.

Here’s a numeric example that makes this clear: a rock song might have 60-120 multitracks in the full session, but only 4-10 stems when delivered for mastering or sync. The entire drum kit folder might contain 14 separate audio tracks (kick in, kick out, snare top, snare bottom, hi-hat, 3 toms, 2 overheads, 2 rooms, a parallel bus)—but the drum stem is just one or two files that capture all of that, already mixed.

Format	What it contains	Typical file count	Processing level
Tracks	Single audio elements	60-150+ per song	None to light
Multitracks	All individual tracks	60-150+ per song	Usually minimal
Stems	Grouped submixes	4-12 per song	Fully processed

A word of caution: Some clients casually say “stems” when they really mean multitracks. This miscommunication can cause delays and extra costs. Always clarify what’s actually needed before you start exporting.

Key differences: stems vs. multitracks in real projects

The distinction matters because stems and multitracks serve different purposes in real-world workflows.

Multitracks are what a mix engineer needs when they’re building a mix from scratch. They want maximum control over every mic and instrument—the ability to adjust the snare bottom mic independently from the snare top, or to process a vocal double differently from the main take. Multitracks give complete creative control.

Stems are what’s requested when detailed remixing of every individual mic isn’t required. This includes:

• Mastering: A mastering engineer in 2025 might request 4-8 stems (drums, bass, music, vocals) from a producer to make targeted adjustments without needing the full session
• Remixes: A remixer needs access stems for vocals and key elements, not 100+ tracks they’ll never use
• Film/TV sync: Music supervisors want to lower drums during dialogue or mute vocals entirely
• Live playback: Touring bands need separate stems they can mute or adjust depending on who’s performing (split songs into 4- or 6-stems)

Stems are usually fully processed and balanced—the mix engineer’s decisions are printed into them. Multitracks may be completely dry, with no bus processing, send FX, or automation applied.

What are stems used for in modern production?

Stems are fundamentally a workflow tool. They speed up editing, enable alternate versions, help with sync licensing, and simplify collaboration across studios and time zones.

The 2010s and 2020s saw a massive shift toward remote collaboration and online content creation. Spotify, Netflix, YouTube, TikTok—all of these platforms drove demand for flexible audio that could be quickly adapted without reopening complex sessions.

Here are the major use cases you’ll encounter:

• Mixing and mastering support (targeted adjustments without full multitracks)
• Remixes and DJ sets (isolating vocals, drums, or hooks)
• Sync for film, TV, streaming, and games (editing music around dialogue)
• Live shows and touring (running backing tracks alongside live instruments)
• Podcast and video editing (separating dialogue from music beds)
• Immersive formats like Dolby Atmos (repositioning elements in 3D space)

Radio edits, clean versions & alternate mixes

Labels and distributors regularly need multiple versions of the same song: radio edits, clean versions with explicit lyrics removed, shorter cuts for playlists, and extended versions for DJ use.

Working from stems makes this dramatically faster than rebuilding from the original mix. Need to create a clean version? Mute the vocal stem during the explicit words and drop in a censored alternative—without touching the 90+ tracks in your session.

Consider this scenario: you need to edit a 3:45 single down to a 3:00 radio edit for a 2025 playlist placement. With stems, you can rearrange sections in minutes. Cut the second verse by working with just 8 stem files instead of managing 100+ individual tracks with all their routing and processing.

Time and cost savings vs. rebuilding the full mix:

• Editing stems: 30 minutes to 2 hours
• Reopening full multitrack session: 4-8+ hours
• No risk of losing plugin settings or mix bus configurations
• Easy to create multiple versions in one sitting

Ensuring the correct versions are created during the mastering process can save significant time and money, as this step is distinct from activities done during mixing or stem editing.

Remixes, DJ sets and creative reworks

Remixers and DJs prefer stems because they can isolate vocals, drums, or hooks without needing access to the full multitrack project or the producer’s original DAW session.

The typical EDM and pop workflow works like this: labels release “remix packs” that include stems—usually vocals, drums, bass, and synths—to producers worldwide. These stem files give remixers everything they need to create new versions while maintaining sound quality.

For live performances, DJs use stems to create mashups and creative transitions. Picture a 2026 club or festival performance where a DJ plays an acapella stem from one track over the drum stem of another. This kind of live playback manipulation is only possible when you have separate stems rather than a fully mastered stereo file.

The quality difference matters too. Trying to extract elements from a finished master using AI separation tools creates artifacts and phase issues. Original stems maintain the full fidelity of the production.

Sync for film, TV, streaming and games

Music supervisors and editors working on Netflix series, YouTube channels, advertisements, and video games frequently request stems as part of any sync deal.

The reasons are practical:

• Muting vocals: Lyrics often clash with dialogue in scenes. Having a separate vocal stem lets editors drop the singing instantly.
• Adjusting drums: Action scenes need punch; quiet moments need subtlety. Lowering or raising the drum stem solves this.
• Looping sections: A tension stem can loop under a cliffhanger without obvious repeats if it’s separated from melodic elements.
• Creating variants: Same song, different energy levels for different scenes.

For example, a 2023 streaming series might need an instrumental-only version for dialogue-heavy scenes and a “no drums” version for intimate moments—all from the same track. With stems, the music editor creates these in minutes.

Many production music libraries launched in the 2010s and 2020s now provide stems by default alongside their full mixes. If you’re licensing music for video, check whether your library includes stems in the download—it can save significant editing time.

Live playback, touring and worship setups

Modern live shows across pop, rock, worship, and electronic music rely heavily on backing tracks. Bands run these from Ableton Live, MainStage, or similar software, synced to a click track for the drummer.

Stems allow musical directors to mute or rebalance groups depending on who’s actually on stage that night. No keys player available for tonight’s show? Keep the keys stem up. Drummer can cover all the percussion? Mute the percussion stem and let them play live.

Here’s a practical example: a 5-piece band on a 2026 tour runs stems for extra percussion, synth pads, background vocals, and FX alongside a click track. Each stem can be assigned to a different output, giving the front-of-house engineer control over backing track levels independently from live instruments.

This approach keeps shows consistent across venues without needing massive live rigs to recreate every sound from the studio. The stems capture the production value; the band adds the live energy.

Immersive, Atmos & remastering older catalogs

Engineers creating Dolby Atmos or surround sound mixes often start from stems—especially when working with older albums from the 1980s, 1990s, or 2000s where original multitracks are incomplete, degraded, or lost entirely.

Stems let engineers reposition major elements (vocals, drums, strings, guitars) in a 3D audio field even when every individual mic isn’t available. The drum stem can move overhead; the vocal stem can sit front and center; ambient instruments can fill the rear channels.

This has become a significant trend in the early-to-mid 2020s, with classic albums and film scores being remixed into Atmos for Apple Music, TIDAL, and Blu-ray releases. Artists and labels are revisiting their catalogs, and stems provide the flexibility needed without requiring every original track.

Stems in the Recording Studio

In the recording studio, stems are an essential tool for streamlining the music production process. When working on a song, engineers often record multiple individual tracks for each instrument—think of separate tracks for the kick, snare, hi-hats, and overheads in a drum kit. To simplify mixing and provide more flexibility, these tracks are combined into a single stereo file known as a drum stem. This drum stem captures the sound of the entire drum kit, allowing the mix engineer to process and balance it as one cohesive unit.

The same approach applies to other elements of a song. A bass stem might merge several bass tracks, while a vocal stem could include all lead and background vocals. By grouping these individual elements into stems, producers and engineers can focus on the overall sound and balance of the mix, rather than getting bogged down in the details of each track.

Stems are typically exported as high-quality WAV files or other audio file formats, making it easy to share them with other artists, producers, or mix engineers. This not only speeds up the workflow in the recording studio but also ensures that everyone involved in the project has access to the key components of the song. Whether you’re working on a pop hit or an indie project, using stems helps you achieve a polished, professional mix while keeping the creative process efficient and collaborative.

How to create stems in your DAW

Every major DAW—Ableton Live, Logic Pro, Pro Tools, Cubase, Reaper, FL Studio—can export stems, but the specific method differs between platforms.

The general workflow follows these steps:

1. Group or bus related tracks together (all drums to a “Drums” bus, all vocals to a “Vocals” bus)
2. Route them through aux/submix channels
3. Export or bounce those buses over the same time range
4. Verify all stems start at exactly the same timecode

That last point is critical. Every stem must begin at exactly the same point—typically 0:00:00 or bar 1, beat 1—so they line up perfectly when imported into another session. Even a few milliseconds of offset creates phase issues and timing problems.

Recommended technical settings:

• Format: WAV (not MP3 or AAC)
• Bit depth: 24-bit (or 32-bit float if your collaborator requests it)
• Sample rate: Match your project (44.1kHz, 48kHz, or 96kHz)
• Consider real-time vs. offline bounce depending on plugin compatibility

Different DAWs, different export methods

Logic Pro workflow:

1. Route your drum tracks to a “Drums” bus using sends or output routing
2. Repeat for bass, guitars, keys, vocals, etc.
3. Solo the bus you want to export (or use the bounce settings to select it)
4. Go to File → Export → 1 Track as Audio File
5. Set your format, bit depth, and sample rate
6. Name the file clearly (e.g., “SongTitle_DrumsStem.wav”)
7. Repeat for each stem group

Pro Tools workflow:

1. Create aux tracks to serve as stem buses
2. Route your track groups to these aux tracks
3. Use Bounce Mix to bounce each aux bus individually, or
4. Use Track Commit to freeze multiple stems at once
5. Ensure your bounce range starts at the session start for alignment

Ableton Live workflow: See what stems are in music production to understand how they can improve and streamline your workflow.

1. Group related tracks using Cmd/Ctrl+G
2. Go to File → Export Audio/Video
3. In the Rendered Track dropdown, select “All Individual Tracks” or choose specific groups
4. Set your export format and bit depth
5. Ableton can export multiple stems simultaneously, saving significant time

Some DAWs offer batch export features that bounce all your buses at once. This is faster than the solo/mute manual approach and reduces the chance of human error.

Grouping choices: how many stems do you really need?

The number of stems you create depends entirely on how they’ll be used.

Minimal approach (4-6 stems): Best for mastering, where the engineer just needs broad control. Example: drums, bass, music, vocals.

Medium approach (8-10 stems): Works for most sync and remix scenarios. Example: drums, percussion, bass, guitars, keys/synths, lead vocal, backing vocals, FX.

Detailed approach (12+ stems): Required for complex remixes or film scoring with specific needs. Example: separating four tracks of synths, splitting acoustic and electric guitars, isolating effects stems.

For a typical pop track delivery, 8 stems covers most needs:

• Drums
• Percussion
• Bass
• Guitars
• Keys/Synths
• Lead Vocal
• Backing Vocals
• FX/Atmos

Pro tip: Maintain consistency across an album or EP. Decide your stem layout once and reuse it for all songs in a 2024-2026 release. This makes life easier for everyone receiving stems downstream.

Wet vs. dry stems (handling effects and processing)

“Wet” stems have reverb, delay, and bus FX printed directly into the audio. “Dry” stems are more raw, leaving room for additional processing later.

Typical delivery includes fully wet stems—your instrument processing is printed in, your vocal reverb is baked in, your drum bus compression is part of the file. This preserves your mix decisions.

However, some situations call for additional flexibility. A remixer in another country might want both:

• A dry vocal stem (just the processed vocal, no reverb or delay)
• A separate vocal FX stem (just the reverb and delay, for re-balancing space and ambience)

Once reverb or delay is printed into a stem, you can’t remove it. If a client specifically asks for drier stems, discuss expectations before you export. Some prefer having more options; others want your complete creative vision intact.

Phase coherence and quality checks

After exporting stems, always verify your work. Import all stems into a fresh session, line them up from bar 1, and play them back. They should sound identical (or extremely close) to your original mix. For more ways to enhance your music production skills, check out this comprehensive guide.

Watch for these common issues:

• Latency from plugins: Some plugins add latency that can cause stems not to line up perfectly
• Linear-phase EQs: These can introduce pre-ringing that affects how stems sum
• Lookahead limiters: May cause timing shifts if not compensated properly
• Stereo image collapse: Check that your stereo spread sounds correct when stems combine

Focus on transients—kick and snare hits should sound tight, not flamming or doubled. A mastering engineer or sync editor relies on phase-coherent stems to avoid weird artifacts when they’re making adjustments.

How to get stems from existing music

The most straightforward way to get stems is to ask the original producer, label, or music library—assuming you have legal rights and a clear license for your intended use.

Professional stem delivery typically looks like this:

• Zipped folder with clearly named WAV files
• Naming convention like “SongTitle_DrumsStem.wav”, “SongTitle_VoxStem.wav”
• README or text file with tempo, key, sample rate, and session notes
• Reference stereo mix included for comparison

Music libraries and subscription services that cater to video and podcast production increasingly provide stems alongside full mixes. If you’re licensing for YouTube video or commercial use, check whether stems are included in your license before you download.

For major-label releases, stems are tightly controlled. Getting them for official remixes typically requires direct contact with the label, publisher, or artist management. This is where having an established relationship (or a legitimate business reason like a signed remix or sync deal) makes all the difference.

Stem separation tools and limitations

AI-based stem separation has exploded in recent years. Tools built on technologies similar to iZotope RX Music Rebalance, LALAL.AI, and Demucs can approximate stems from a finished stereo master when original stems are unavailable.

The pros:

• Quick extraction of vocal, drum, bass, and instrumental elements
• Useful for practice, unofficial remixes, or restoration projects
• Accessible to anyone with a computer and the right software

The cons:

• Artifacts and audio degradation, especially in complex mixes
• Phase issues when recombining separated elements
• Legal concerns if used without permission for commercial releases
• Quality doesn’t match stems exported from the original session; to learn more about improving sound quality in your music production process, consider the benefits of automated mastering.

For professional commercial work in 2024-2026, original exported stems remain the gold standard. AI separation is improving rapidly, but it’s still a compromise compared to stems that came directly from the producer’s session.

If you’re considering AI separation for a commercial project, consult with the rights holders first. Using separated stems without permission can create legal exposure, even if the technology makes it technically possible.

The Role of a Mastering Engineer

The mastering engineer is the final gatekeeper in the music production process, responsible for preparing the final mix for release across all platforms. Stems are a powerful tool in the mastering engineer’s toolkit, offering the ability to make precise adjustments to specific elements of a track without altering the entire mix.

When a mastering engineer receives stems—such as a bass stem or a vocal stem—from the mix engineer, they can address issues or enhance certain aspects of the music with greater accuracy. For example, if the bass needs more presence or the vocals require a subtle boost, the mastering engineer can work directly with those stems to achieve the desired result. This level of control ensures that the final mix sounds balanced and professional, no matter where it’s played.

Using stems in mastering also allows for creative problem-solving. If a song’s low end is overpowering, the engineer can adjust just the bass stem. If the vocals need to sit better in the mix, the vocal stem can be fine-tuned independently. This targeted approach leads to a more polished and impactful final product, helping the music stand out on streaming platforms, radio, and beyond.

Best practices for naming, organizing and delivering stems

Clean organization saves hours on large projects, especially when you’re collaborating with studios across time zones. A mastering engineer in London doesn’t want to guess which file is which when they open your folder at 2 AM their time.

Recommended naming convention:

SongTitle_Artist_StemType_Version.wav

Example: “MidnightCity_2025_DrumsStem_v1.wav”

Folder structure:

/MidnightCity_Stems/
  MidnightCity_DrumsStem.wav
  MidnightCity_BassStem.wav
  MidnightCity_GuitarsStem.wav
  MidnightCity_SynthsStem.wav
  MidnightCity_LeadVoxStem.wav
  MidnightCity_BVsStem.wav
  MidnightCity_FXStem.wav
  MidnightCity_StereoRef.wav
  README.txt

What to include in your README:

• Sample rate and bit depth
• Tempo (BPM) and key
• Any bus processing notes
• Contact info for questions

Common labeling conventions:

• “INST” = instrumental (no vocals)
• “NO VOX” = alternate without lead vocal
• “TV MIX” = broadcast version (often dialogue-friendly)
• “CLEAN” = explicit content removed

Always include a reference stereo mix. This lets the receiving engineer compare their work against your intended balance and catch any obvious issues.

Communicating with clients and collaborators

Clear communication prevents the single most common stem-related problem: someone asking for “stems” when they actually mean full multitracks.

Always clarify terminology upfront. Before you start exporting, send a quick message asking:

• “Do you need full multitracks, or 4-8 grouped stems that recreate my mix?”
• “Wet stems (with reverb/delay printed) or dry stems?”
• “What format? 44.1kHz or 48kHz? 24-bit or 32-bit float?”
• “Any specific stem groupings you require?”
• “What’s your deadline?”

This five-minute conversation prevents hours of rework. Clients appreciate the professionalism, and you avoid the frustration of delivering the wrong files.

Why stems matter for your long-term catalog

Archiving stems is a safeguard for your future. They’re smaller than full multitrack sessions but far more flexible than a single stereo master file.

Think about where your music might go in 5, 10, or 15 years:

• A 2012 release gets picked up for a 2025 Netflix sync opportunity—and they need an instrumental version
• Your album gets selected for a Dolby Atmos remix in 2027
• A label wants to release a deluxe reissue with alternate versions
• A remixer from a genre you’ve never heard of wants to rework your track

If all you have is a stereo master, your options are limited. If you have stems, you have flexibility.

Simple archival workflow:

• Back up stems to at least two physical drives in different locations
• Use a reputable cloud service as a third backup
• Organize by year and project: /2026/AlbumTitle/SongName_Stems/
• Update backups after any new versions or exports

Losing stems can directly limit future revenue. Sync deals, game placements, immersive releases, and remixes all become harder—or impossible—without access to separated elements. Given how often catalog music gets rediscovered on streaming platforms, protecting your stems is protecting your business.

Conclusion and Future of Stems

Music stems have become an indispensable part of music production, offering artists, producers, and engineers greater creative control, flexibility, and opportunities for collaboration. As technology continues to advance, the role of stems is only set to grow—especially with the rise of immersive audio formats like surround sound and Dolby Atmos, where having access to separate stems is crucial for creating dynamic, engaging listening experiences.

Looking ahead, stems will continue to empower producers and artists to push the boundaries of what’s possible in music. Whether you’re remixing a track, preparing your music for a sync deal, or exploring new audio formats, working with stems ensures you have the control and versatility needed to adapt to any creative challenge. By embracing stems as a core part of your workflow, you’ll be ready to take your music to new heights and stay ahead in the ever-evolving world of music production.

Summary: what stems are and how to use them effectively

Stems are grouped, processed audio files that act as mini-mixes of your drums, vocals, instruments, and effects. When combined, they recreate your complete song—or at least come close enough that a mastering engineer, sync editor, or remixer can work with them effectively.

The key distinctions to remember:

Format	Description	Use case
Single track	One audio element	Building a mix
Multitracks	All individual tracks, often raw	Mix engineer needs full control
Stems	Grouped, processed submixes	Mastering, remixes, sync, live

Core benefits of working with stems:

• Faster edits and alternate versions without reopening complex sessions
• Simpler collaboration across studios, countries, and time zones
• Flexible sync and live playback setups
• Easier mastering adjustments
• Future-proofing your catalog for formats that don’t exist yet

Whether you’re a producer, artist, video editor, or podcaster, establishing a stem workflow now pays dividends later. Don’t wait until a client demands stems—build the export process into your finishing routine on your next project.

Create your stems. Organize them clearly. Back them up. Your future self (and your future collaborators) will thank you.