The DeliveryLeak Report
A study we intend to run: instrument real Indian weddings, end to end, and find out how many guests actually receive a photograph of themselves, how many reshare it, whether the studio’s name survives the forward, and how many of those guests ever become an enquiry.
Not run. Not published. There are no findings on this page. We are designing the study and looking for studios willing to have their events instrumented. Everything below describes what we intend to measure and how — nothing below describes a result.
Both links go to our contact form. There is no mailing list to join yet.
Nobody can tell you what happens to your work after you deliver it.
A photographer delivers three thousand frames from a wedding. Several hundred people appear in them. Some number of those people will see a photo of themselves; some number of those will send it onward; some smaller number will one day need a photographer. Ask any studio in the country how big those numbers are and you will get a guess, delivered with total confidence, that contradicts the guess of the studio next door.
We have looked for a published measurement of this — for the Indian wedding market or any other — and have not found one. That may be a failure of our searching rather than a gap in the literature. If a credible study already exists, send it to us: we would rather link to it than spend a year reproducing it.
Your work circulating unbranded, untracked and unattributed. The core disease of the delivery category.
We named the leak before we could size it, which is the wrong order and we know it. This study is the attempt to put a measured number behind a term we have been using argumentatively. Until it exists, treat the Delivery Leak as a hypothesis — a well-motivated one, but a hypothesis.
Four questions, no answers yet.
The report is designed around four measurements. Each is stated as a question because that is genuinely all we have.
Reach
Of the people physically present at an event and visible in the photographs, what share ever receives a single photo of themselves? Not the couple — the guests.
Reshare
Of the guests who do receive a photo, what share reshares it — to a story, a status, a family group — and over what window does that resharing decay?
Attribution survival
When a photo is reshared, how often does anything identifying the studio survive the forward, the screenshot and the recompression?
Conversion
What share of reached guests becomes a named, consented enquiry to the studio — the Guest-to-Lead Rate — and how long after the event does it arrive?
How we intend to run it.
Published in advance so that it can be criticised in advance. If something here is methodologically weak, we would much rather hear it now than after we have spent a season collecting data with it.
Recruit studios, not a lab
Real weddings shot by working Indian studios on their normal schedule, across a spread of city sizes, budgets and guest counts. No staged events. Studios can withdraw their data at any point before publication.
Instrument the delivery moment
Measure what happens between handover and forgetting: how many guests arrive, how many find themselves, what leaves the gallery and where it goes. Several of the instruments this requires are still in development — see below.
Collect consent separately, twice
Under DPDP purpose limitation, the consent a guest gives to find her own photos does not cover being counted in a study. Study participation will be its own explicit, plainly worded, easy-to-decline opt-in, and non-participation must cost a guest nothing.
Publish the method before the numbers
The instrument, the definitions and the exclusion rules go out first, so the findings can be argued with rather than merely quoted. Aggregates only — no guest-level data, no studio named without written permission.
Publish it even if it kills our argument
If it turns out guests already receive their photos, already reshare with attribution intact, and already convert, then Delivery-Led Growth is a solution to a problem that does not exist and we will say so on this page.
The study needs instruments we have not finished building.
This is the honest reason there is no report yet. Measuring reshares, consented opt-ins and guest-to-lead conversion requires an attribution layer that does not exist in Studioshare today. Every item below is in development, not available, and the report cannot be run properly until enough of it is real.
Guest Graph consent capture
A separate, explicit, easy-to-decline opt-in that lets a guest hear from you later — kept legally distinct from the consent that powers "find my photos".
Earned Impressions tracking
Measure how often guests reshare your branded work, so reach stops being a guess.
Guest-to-Lead dashboard
Your G2L rate per event: how many guests arrived, how many opted in, how many became real enquiries.
Second-order booking attribution
Trace a signed booking back to the event where that client first showed up as somebody else’s guest.
macOS and iOS apps
Android is live on Google Play and the desktop capture app runs on Windows. macOS and iOS are still in development.
Windows desktop app and Android. macOS and iOS in development. Face matching runs on our servers — a 512-dimension AuraFace embedding matched by cosine nearest-neighbour search — not on the guest’s device.
What this page deliberately does not contain.
- Not included: Findings. There are none. Nothing has been collected and nothing has been analysed.
- Not included: A preview statistic, a teaser percentage, or an “early signal” from a pilot.
- Not included: A publication date. We are not announcing one we cannot hold.
- Not included: A sample size, a study partner, or an institutional affiliation. None are confirmed.
If you see a number attributed to The Delivery Leak Report anywhere — in a pitch deck, a blog post, or one of our own sales conversations — it did not come from us, because we have not produced one. This page is the canonical record of what has and has not been published.
Two ways to be involved.
Put your events in the study
If you shoot weddings in India and want to know what actually happens to your work after handover, tell us. We will explain exactly what would be measured, what your guests would be asked, and what you would get back — before you commit to anything.
Be told when it publishes
Send us a line through the contact form and say you want the report. That is the whole mechanism — there is no signup form here yet, because we would rather not run a capture form for something that does not exist.
In the meantime, the argument this study is meant to test is written out in full elsewhere: the manifesto is why we think delivery is being wasted, and the framework is the definition, the language and the metric.