Fifty products should look like one shop

One good product photo is judged on its own page. A catalogue is judged somewhere else: on the collection page, the search results and the grid of thumbnails a shopper scrolls before opening anything. There, every image is seen next to its neighbours, and the thing a shopper notices first is not whether one shot is beautiful. It is whether the shop looks like one shop.

That changes what a bulk run is for. It is not fifty separate shoots that happen to share a button. It is one decision about how the collection should look, applied fifty times. Most of the work is making that decision before you press start.

Decide what must match, then split the run by it

A Nitaris bulk run takes up to fifty products and applies one shot count and one shape to all of them. On the web it also takes one style note, a line like "warm, editorial, natural light", for the whole run. Those settings are the collection's look, so choose them for where the collection will be seen rather than for any single product.

If two parts of your catalogue need different looks, that is two runs, not one. Rings and necklaces that sit in the same grid should share a shape. A homeware range and a jewellery range probably should not share a style note.

Name the files before you drop them in

On the web you drop a folder of photos, and Nitaris guesses which files are the same product from their names. The rule is deliberately narrow, and knowing it saves a lot of fixing:

• A trailing number joins files. ring-01, ring-02 and ring (3) become one product called Ring.
• A trailing word never does. ring-front and ring-back stay two products, because ring-gold and ring-silver look exactly the same to a filename, and merging two products into one shoot is the expensive mistake.
• Camera names are a trap. IMG_4021 and IMG_4022 reduce to the same stem. A whole folder of them is split back into one product per photo, but two to five of them will be joined into one product, whether or not they are.

So rename first: the product's name, a hyphen, a number. Then read the grouping screen before you start. You can rename, merge and split products there, and nothing is charged until you press start. In the Shopify app there is nothing to name: you pick the products, and their photos come from the listing.

Keep the source photos alike

Every set is generated from that product's own photo, which the render is built to stay faithful to. That is the point of it, and it has a consequence for a catalogue: a collection photographed on the same table, from a similar distance, at a similar angle, gives the run a far more even starting point than one assembled from supplier shots, phone photos and last year's studio day. You do not need a studio for this. You need the same spot by the same window, and ten minutes with each product.

If you use on-model shots, give the collection one model. Each product row has its own model picker, and the web uploader pre-fills every row with the model you use most, so the default is already consistent. Changing one row is how a single product ends up with a different face from the rest of the grid.

Run three before you run fifty

Before committing a whole collection, run three products that are as different from each other as the collection gets: the smallest, the shiniest and the one with the most text on it. Put their hero shots side by side. If those three look like one shop, the other forty-seven usually will too, because the settings that make them match are the same settings. If they do not, you have found out for the price of three sets instead of fifty.

Standard or overnight

A bulk run offers two lanes, and the choice is only about when you need the photos:

• Standard renders the run straight away. Use it when the photos are needed today.
• Overnight returns within 24 hours and costs fewer credits: a six-shot set is four credits instead of six, and a three-shot set is two instead of three. There is no promised time inside the 24 hours, and a set that does not come back is not charged. Use it for anything planned: a launch next week, a seasonal refresh, a backlog you have been meaning to reshoot.

If some products fail, you can retry only those, and switch lanes for the retry.

Check the grid, not the files

When the run finishes, do not review it product by product. Look at it the way a shopper will: the hero shots in a grid. The outlier is obvious there and invisible anywhere else. It might be a darker set, a tighter crop, or a lifestyle frame where every other product got a plain background. Each product in a run becomes an ordinary shoot in your library, so an outlier can be reshot on its own without touching the other forty-nine.

Then check each product the way you would check any render, before it goes on a listing. An earlier post, "check the cap before the colour", lists the order to do that in.

Films are a separate choice

A bulk run makes photos, not films. That is a reasonable split rather than a gap: a catalogue needs every product photographed, but it rarely needs every product filmed. Pick the handful whose look is the reason people buy them, the hero of the collection and the best sellers, and make a film of each on its own.

The short version

Choose one shape and one shot count per collection, and split runs by collection. Rename files to product-name-number before uploading, and read the grouping screen. Photograph the source photos the same way. Keep one model. Run three products first, compare their hero shots, then run the rest overnight unless you need them today. Review the finished run as a grid, and reshoot the outlier on its own.