Stop delivering photos.Delivery is the channel.

The category taught you that delivery is a cost.

Kwikpic. FotoOwl. Memzo. AccioPix. Pixieset. Pic-Time. Six products, six pricing pages, one shared premise underneath all of them: that a photographer’s job ends when the files reach the client, and that everything between the shutter and the handover is overhead to be squeezed.

You can read the premise straight off what the category benchmarks itself on. Upload speed. Match accuracy. Cost per photo. Storage included. Not one of those is a business outcome. They are logistics metrics, every one of them, and a category that competes on logistics metrics has exactly one destination: the price floor.

It is already arriving. Our read of where per-photo face matching is heading is paise — we have no survey to hand you, and you should treat that as a direction rather than a quoted price. But once the thing you sell is measured that way, “we deliver faster” is not a business — it is a feature, waiting to be absorbed by whoever has more servers and less patience for margin.

Look at what you actually own at two in the morning.

Four hundred people were in that hall. Among them: the bride’s colleague who is getting married in November and has not booked anyone. Two cousins who will post to their stories before you have finished packing the lights. An aunt who will screenshot the group shot and send it to a WhatsApp group of eighty people you will never meet. A father who spent the evening watching how you handled his family and has a second daughter.

That is a warm, self-selecting, geographically concentrated audience carrying live intent, assembled at somebody else’s expense, watching your work on purpose. A consumer brand would pay heavily to stand in front of a room like that for ten minutes. You stood in front of it for fourteen hours, and they liked you.

Then the gallery link goes out, and the entire thing evaporates. Not slowly — within days. By the time the couple is back from their honeymoon, the audience you built has dispersed into four hundred camera rolls with nothing connecting any of them back to you.

The problem is not reach. The reach is enormous. It is invisible.

This is the part that gets misdiagnosed. Photographers are told they need to post more, run more ads, be more consistent on Instagram — as though the shortage were attention. There is no shortage of attention. Every event you shoot throws off more genuine, earned, human distribution than a month of paid social, and it happens whether you participate or not.

The shortage is attribution. The work travels; the credit does not. Forwards strip metadata. Screenshots strip watermarks. A folder link tells you nothing about who opened it, who saved from it, who sent it onward, or who would have called you if there had been anything to call. We call that leakage the Delivery Leak, and it is the core disease of this category.

We should be precise about who is guilty here, because it would be dishonest to exempt ourselves. Dark delivery — real reach, zero attribution — is what every platform in this category ships today, including Studioshare. The difference we are claiming is not that we solved it. It is that we think it is the problem worth solving, and the rest of the category thinks the problem is the invoice.

Delivery is the acquisition channel. It always was.

Product-Led Growth was a useful idea because it moved a line item. It said the product is not merely what you sell after marketing does its job — the product is how you acquire. Once a company believed that, it stopped funding the product out of the cost budget and started funding it out of the growth budget, and everything about how it was built changed.

Delivery-Led Growth says exactly the same thing about the delivery moment. The forty-eight hours after an event, when guests are actively hunting for their own face and are one tap from a story post, are not the end of the job. They are the top of the funnel. They are the only top of funnel in this business that you never have to buy, because you already paid for it with a night of your life.

Set the two side by side. A photographer spends the month’s ad budget renting attention from strangers who have never seen the work, in a feed, against everyone else’s ad budget. The same photographer, in the same month, hands several hundred people who watched them work all night an unbranded link, and asks for nothing. Then wonders why growth feels like pushing.

Change the denominator and the whole business changes.

Cost per photo is a denominator that can only ever go down, and every rupee you win there is a rupee, once. Cost per booking is a denominator that compounds, because the same event can produce a booking this year and another one in three years when the guest who found herself in frame 2,411 gets engaged.

That is what the metric we care about is for. The Guest-to-Lead Rate asks a question no delivery platform currently answers: of the guests at that event, what share became a real enquiry? Underneath it sit earned impressions — reshares of your branded work that you did not pay for — and second-order bookings, the ones traceable back to an event where the client first showed up as somebody else’s guest.

None of these are exotic. Every other channel a business uses has been measured this way for thirty years. Delivery is the one place where an entire industry agreed not to look.

Consent to find your photos is not consent to be marketed to.

There is an obvious cheap version of everything written above, and we want to name it before somebody builds it. Take the selfie a guest submits to find her own photos, keep the identity, keep the number, and start sending things. It would work for about a year.

Under India’s DPDP Act, purpose limitation means the permission a guest grants to find her own face does not extend to hearing from you afterwards. Those are two different permissions and they have to be collected twice. So the Guest Graph — the consented identity map an event builds — will only ever be assembled from a second, explicit, plainly worded opt-in that is easy to decline.

This is slower. It converts worse than not asking. It is precisely the part that a competitor racing to the price floor will skip, and skipping it will look like an advantage right up until the first school administration or corporate legal team reads the flow. We would rather be the version that survives that meeting.

What we have built, and what we have not.

A manifesto is cheap. Anyone can claim a category. The honest thing is to put the argument next to the software and let you see the gap, so here is the gap, with nothing rounded in our favour.

Two details we say plainly because the flattering version is available and we are not taking it. The tethered capture app runs on Windows, and the Android app is live on Google Play; macOS and iOS are in development, not shipping. And face matching runs on our servers, not on the guest’s phone — a 512-dimension AuraFace embedding computed on Cloud Run and matched by cosine nearest-neighbour search against the event gallery. “On-device” reads better in a feature list. It is not what happens.

Which leaves us publishing a manifesto for a category whose measurement layer we have not finished building. That is an uncomfortable sentence to put on our own website. It is a great deal less uncomfortable than you discovering it after you have paid.

Windows desktop app and Android. macOS and iOS in development.

Your next booking is standing in the room you are packing up.

We are not asking you to believe a statistic. We do not have one. The study that would produce one — instrumenting real weddings to find out how many guests ever receive a photo, how many reshare it, and how many convert — has not been run, by us or by anyone we can find. We have written down what we intend to measure and published that instead.

What we are asking is smaller and harder: stop accepting the premise. Delivery is not the invoice at the end of the job. It is the only moment in your year when the people most likely to hire you are all in one place, looking at your work, feeling something about it. Treat that as a cost centre and you will keep buying strangers. Treat it as a channel and it compounds.

That channel is the category we are naming: Delivery-Led Growth. And the honest footnote to any manifesto is the part that is not finished — the attribution layer this category ultimately rests on. Consented guest identity, countable earned impressions, a Guest-to-Lead Rate you can read off a dashboard: none of it is in the product today, and all of it is still being built. We would rather name the category now, with the gap showing, than wait until it closes and pretend we always had it.