As a hobbyist photographer, summer trips are an opportunity to rethink your lens line-up and decide if it's time to invest in a new one. This is a thrilling yet daunting task, as you quickly get mired into discussions on MTF curves, pixel peeping, sharpness, image quality, autofocus speed and more. What should be a fun decision turns into analysis paralysis—and before long, it stops being fun.

Here’s how I let ChatGPT and a bit of EXIF data make the call for me.

The Grand Canyon (Fuji X-T5 + 18–55mm)

Step 1 - Narrow it down

"I am looking at a lens to buy for my Europe trip. I'm shooting on a Fuji XT-5. I have the 50-140 lens and I love it. I have a fixed 35 mm but I find it too narrow to be useful. I also have the 18-55 lens but I don't use it that much, but can keep it if needed. Can you check prices and availability on https://www.mpb.com and compare to list prices to see what would be a good discount? I'm looking for a wide angle, like Fujifilm XF 23mm or the 16-55 2.8 lens"

I used OpenAI’s o3 model and it came back with an amazing analysis, looking at the type of photography that I will do (landscape, street, family), the lightness of the lens and suitability for travel and the discount I can get on a 2nd hand lens. I asked for tables of weight, percentage-off list and suitability for low light.

Some of the initial analysis

Table: Lenses ordered by weight

After some back and forth I narrowed it down to two prime lenses—18 mm or 23 mm—but the specs were too close to call.

Step 2 - Data Analysis

I have a zoom lens that covers the range from 18mm to 55mm. I don't use it a lot, but I'm curious, when I use it, which range did I prefer. Was it closer to 18mm or to 23mm?

This is when inspiration struck. I asked ChatGPT:

"Can you help me create a script for MacOS that I can run on a folder of photos to analyze the focal lengths I use most and generate a histogram I can open in Excel? Please convert them from iPhone and APS-C format to 35mm format"

After a few back and forth conversations, installing homebrew, exiftool and other libraries, everything looked right. With the script (GitHub) ready, I ran it on my Grand Canyon folder, pulled focal length data, and converted it to full-frame equivalents. Then I plotted a chart to see where I naturally shoot most often.

Chart: Number of photos by focal length

The chart makes it clear: the photos I took tend to cluster mostly around the 18mm range. There was still a healthy spread across focal lengths, which suggests that a zoom lens could be a strong option down the line—but for now, 18mm felt like the right call.

So I went with the 18 mm—and, for once, the journey was as fun as the gear. The histogram backed up the pick, the o3 tables kept me honest on weight and price, and the decision felt data-driven instead of gut-driven. Now all that’s left is to pack my bag and start taking photos.