Share Your Work & Win Prizes - Narrative International Photography Awards 2026

Elli Kim Content
Thursday, August 6, 2026
Lightroom Classic renders previews from RAW files on the fly unless you change how it builds them on import. Three fixes close most of the gap: import with Embedded & Sidecar previews, enable GPU-accelerated preview generation, and raise your Camera Raw Cache. For high-volume shoots, pairing these with a dedicated culling tool cuts pre-edit review time from hours to under an hour.
In this article
The Real Reason Lightroom Feels Slow When Culling
The Fixes That Actually Help (And Their Limits)
Choosing Your Best Shots Faster
Tips for a Faster Photo Editing Workflow
What This Looks Like in Practice
The MacBook Air Question
You just wrapped a full-day shoot. You're home, the hard drive is plugged in, and you open Lightroom. Then you wait. The spinning wheel appears. The grid loads thumbnail by thumbnail. You click the first RAW to zoom in and check focus - and wait again.
If this sounds familiar, you're not imagining it. And the bad news is: your laptop probably isn't the problem.
Lightroom wasn't designed for the culling workflow most photographers actually use. Every time you move between images in the Library module, Lightroom has to generate or retrieve a preview of your RAW file. If those previews haven't been pre-built, it renders them on the fly. It uses your CPU, hits your drive, and stalls your flow.
The preview system is where most photographers feel this most acutely. When you import a 1,000-photo wedding shoot and ask Lightroom to create full-size 1:1 previews upfront, you're telling it to develop 1,000 RAW files at once, hammering your CPU and filling your hard drive. Skip the previews to save time on import, and you pay for it on the other end: every zoom-to-check-focus becomes a loading event.
Lightroom 1.0 shipped in 2007 when modern mirrorless cameras didn't exist. Neither did 45MP RAW files. The gap between what Lightroom was built for and what photographers are asking it to do in 2026 is real, and no amount of cache-tweaking closes it entirely.
This isn't a MacBook Air problem. It's a Lightroom problem.
・・・
There are real tweaks that improve Lightroom's culling performance, and they're worth knowing.
Set Build Previews to Embedded & Sidecar on import. This skips Lightroom's rendering step and shows the JPEG preview your camera already embedded in the RAW file, so import finishes in minutes instead of tying up your CPU. The tradeoff: preview quality depends on your camera's embedded JPEG, and Lightroom still re-renders from the RAW the moment you open Develop.
Build Standard previews on import. This front-loads the rendering process. The initial import takes longer, but browsing becomes faster because Lightroom reads pre-built files instead of generating them from RAW data on the fly.
Increase your Camera Raw Cache, but check your drive headroom first if you're on a Mac. The default 5GB is undersized for modern workflows, and raising it to 50 or 100GB can meaningfully improve Develop module image-switching speeds. Base MacBook Air and Pro configs often start at 256 or 512GB, all fixed at purchase, so a large cache plus a growing preview catalog can eat a real share of your drive. If space is tight, point the cache at an external SSD instead.
Use Smart Previews for editing. These lightweight DNG proxies are a fraction of the original file size, making the Develop module noticeably snappier, especially on a laptop or when working from an external drive.
Keep your catalog lean. Performance starts to degrade once a catalog exceeds around 100,000 photos. Splitting by year or project keeps things responsive.
Buy enough unified memory upfront if you're on Apple Silicon. Unlike a PC, memory on any M-series Mac is soldered at the factory. There's no retrofit later. For a smooth Lightroom experience in 2026, 32GB is the baseline worth paying for at checkout. 16GB introduces measurable UI lag with AI tools and advanced masking, and it's also the minimum threshold for GPU-accelerated preview generation, so an 8GB config misses out on that entirely.
These all help. But here's the honest reality: a photographer shooting 2,000 images at three seconds per image is still looking at nearly two hours of pure decision-making before a single edit begins. Upgrade your RAM, and you've still got two hours of decisions to make.
That's the problem a dedicated culling tool solves.
Lightroom Classic has started closing this gap. Assisted Culling, which reached general availability in version 15.4, analyzes focus, sharpness, and eye state across an import and flags likely rejects. It's a real improvement over fully manual review. But it still runs on top of the same preview pipeline that's slow to begin with, and it only works within Lightroom's own catalog.
Dedicated culling software takes a different approach: it evaluates your entire shoot on import, independent of Lightroom's preview engine. Focus, sharpness, facial expressions, eye openness are assessed across every frame before you've clicked anything, and the results carry over cleanly when you're ready to edit. By the time you start culling, the AI has already done a first pass. Your job becomes confirmation, not discovery.
Narrative is built for exactly this. You can import and browse thousands of images in seconds, fully offline, with no cloud upload required. Similar frames are grouped by scene, so you pick the winner of each set without fatiguing over near-identical shots.
The result isn't a faster Lightroom. It's a workflow where the 1-2 second preview lag disappears entirely, because you're not navigating a RAW library at all.
・・・
The bottleneck after most shoots isn't editing - it's the decisions that come before it. Standing in front of 800 near-identical frames, trying to spot the one where the expression landed perfectly, is mentally expensive work. It's also the part most photographers describe as the least satisfying.
A few things make that process sharper. Shoot with selection in mind: give yourself a beat between setups, so your frames group naturally into scenes rather than blurring into one long sequence. That context matters when you're reviewing later. Resist the urge to review everything at full size on the first pass. A quick run through at fit-to-screen - flagging obvious rejects and clear standouts - is faster than zooming into each frame individually and is usually accurate enough to get from 800 images to 200 in under twenty minutes.
The more durable fix is structural. When similar frames are automatically grouped and ranked by technical quality, the decision changes character. Instead of asking "which of these 40 frames is the sharpest?", you're asking "does the AI's top pick match what I felt in the moment?" That's a faster, more instinct-led question - and it's the one Narrative is built to put in front of you. AI assesses focus, sharpness, and expression across every frame before you've touched anything. You move through the shoot confirming calls, not making them from scratch.
・・・
The editing session itself moves faster when the groundwork is done before Lightroom opens. A few habits compound over time.
Shoot tethered or use in-camera rating to flag standouts on the day - even rough marks cut first-pass review time significantly. On import, apply a base preset so every RAW lands with a consistent starting point rather than a flat profile. Use virtual copies when you want to explore a different treatment, so the original stays untouched and your catalog doesn't bloat with duplicate edits.
Batch-sync your develop settings across similar frames - correcting one backlit reception shot and syncing it across thirty saves more time than any single shortcut. Work in collections rather than the full catalog view; a smaller, curated set keeps your attention on the images rather than the library. And resist switching between culling and editing mid-session: each mode of thinking slows the other down.
The bigger gain comes from separating culling and editing entirely. When they share the same session, each drags on the other: you're making technical decisions (is this sharp?) at the same time as creative ones (does this frame tell the story?). Split them into distinct passes and both go faster. Culling is quicker without the temptation to start grading. Editing is cleaner when every image in the session has already earned its place.
A typical 2,000-image wedding shoot in Lightroom, starting from scratch:
Import + build previews: 30-60 minutes
Culling (by hand, Library module): 2-4 hours
Total before a single edit: up to 5 hours
The same shoot with Narrative:
Import + AI assessment: instant
Culling (reviewing AI-ranked tiers, confirming picks): 30-60 minutes
Ship selected images to Lightroom with your personal presets applied: one click
Your Lightroom session starts where it should - at the edit, not the cull.

・・・
"Best culling app for MacBook Air" is a common search for a reason.
Apple Silicon Macs are genuinely powerful machines, and Lightroom Classic (14.5 and later) can now offload preview generation to the GPU, cutting the CPU load that used to cause thermal throttling on long sessions. But GPU acceleration only speeds up the rendering step itself. It doesn't touch the actual bottleneck: building thousands of previews on import and clicking through a Library grid one photo at a time. Even with GPU acceleration on, a MacBook Air still processes every image in sequence.
This applies more to the Air than the Pro. The Pro's active cooling lets its M-series chip sustain GPU-accelerated preview generation through a long import session without throttling. The Air's fanless design means performance can dip during back-to-back culling sessions, especially in a warm venue.
Narrative is fully optimised for Apple Silicon and processes entirely offline. For MacBook Air users in particular, moving culling out of Lightroom is the single highest-impact change you can make to your editing speed.
Try Narrative free on the Ultra plan - no credit card required.
Import your next shoot, let the AI do its first pass, and see what your culling time actually looks like when the bottleneck is removed.
Content
Subscribe to get our productivity blog sent straight to your inbox – and level up your business.