The Age ofCustom Softwarefor Businesses
Independent businesses are no longer limited to the software they can buy.
They build their own, and they run on it.
sampled, live apps
The report's working universe: a sample, not the whole.
industries represented
From healthcare and retail to poultry farming.
traffic on sampled apps
Real customers, clients, and staff.
Businesses have always adapted to the software available to them. Now, they are adapting the software to themselves.
We focus only on deployed apps because a live product is evidence. It tells us a business trusted what it built enough to run on it.
We then spoke to the operators behind these apps. Different industries. Different businesses. But their stories almost always began the same way: something in the business had quietly stopped working.
See Methodology.
WHO BUILT THEM
Almost none were built by a development team.
They were built by the people who run the business.
The problem
Different businesses, same three failures
Underneath the industries, the answers narrowed to three problems almost every operator recognized.
Efficiency was leaking into manual work.
Businesses kept running, but on manual effort that grew with every customer.
WHAT IT LOOKED LIKE:
- An exam-coaching center managed 4,000 student records by hand
- A booking agency entered jobs for 70 contractors one at a time
The software was costing more than it returned.
The money was leaving anyway, in fees, cuts, and subscriptions that never quite closed the gap.
WHAT IT LOOKED LIKE:
- Home sellers hand agents $20,000-$50,000 in commission to do little more than post a listing
- Monthly payments for tools that fit halfway
- A cut to a middleman sitting between them and their own customers
The software they needed simply did not exist.
Their problem was too specific, their market too small, or their idea too new for any vendor to have built a product for it. So they ran without one.
WHAT IT LOOKED LIKE:
- A problem too specific for any vendor to chase
- A market too small to justify a product
- An idea too new for anything on the shelf
How they coped
Every workaround came at a cost
Only one in seven of the 300 operators interviewed was genuinely well served by software already on the market. Everyone else adapted. They accepted manual work, compromised on software that only half-fit, or paid for custom development. None of those choices solved the underlying problem.
How operators coped before they built
Source: 300 structured operator interviews
Out of reach
Why a real fix stayed out of reach
A real fix meant building custom software. For most operators, that door was effectively closed. The quotes alone ended most conversations.
Typical developer agency quote, basic buildThe median
Quotes reported by operators. $20,000 is the median, not the mean.
Even when the money was there, the calendar was not, with builds running from a few months to well over a year while the problem kept costing them daily. And most of these operators had never written a line of code, so building it themselves was never a real fallback.
Share of operators citing each blocker
Of the 300 operators interviewed. Many hit more than one wall, so shares add past 100%.
- Prohibitive development cost
- Timelines from several months to more than a year
- Tried other approaches, from no-code tools to freelancers, and hit a wall
Source: 300 structured operator interviews
The choice came down to build it or live without it, and for years living without it usually won.
The solution
Building moved to the operator
The people who get to build changed. These systems are now built by the operators themselves: people with no coding background and no developer to hire, who a year earlier had no way to build anything at all. Building stopped being reserved for those who could write code or afford someone who could, and moved to the person who runs the business and knows exactly what it needs.
What they built runs both sides of the business.
FACING THE CUSTOMER
Storefronts, marketplaces, booking pages
The sites that bring leads in.
FACING THE CUSTOMER
BEHIND THE COUNTER
CRM, inventory, client portals, internal tools
The machinery no vendor ships in the shape a specific business needs.
BEHIND THE COUNTER
The portraits below cover the industries building at the highest volume. Two operator prompts each, in their own words, show the range within a single vertical: find yours, and see what people in it actually built.
Click an industry to read operator prompts, verbatim.
Running on it
These apps became a part of the business
These are not prototypes. They generate revenue and sit at the operational core of the business. Most are not single-purpose tools either. They bring together the services a business needs into one working system.
Every app sits on a database by default, so the question is what operators wired in on top. Looking across the apps they built, most wired in something real.
What counts as an integration
An integration is an outside service wired into the app: the payment rail that takes the money, the email that sends the receipt, the maps layer that finds the customer. These are the brands operators wire in most.
Stripe
Razorpay
PayPal
MongoDB
Supabase
WhatsApp
Resend
Google Maps
Anthropic
Gemini
ElevenLabsadded at least one integration beyond the database
added two or more: a payment rail, an email or messaging service, a maps or storage layer, composed into one system
Source: Emergent platform data, deployed-app cohort
The connections land where money comes in and customers are served:
take payment
The app generates revenue directly.
25% OF DEPLOYED APPS
send transactional email
Receipts, confirmations: the messages a business sends without anyone touching them.
33% OF DEPLOYED APPS
take payment
The app generates revenue directly.
25% OF DEPLOYED APPS
send transactional email
Receipts, confirmations: the messages a business sends without anyone touching them.
33% OF DEPLOYED APPS
You do not wire a checkout or an automated receipt into something you are experimenting with.
Two behaviors outside the codebase say the same thing:
run on a custom domain
Pointed at real customers.
33% OF DEPLOYED APPS
kept building after launch
Came back to keep building a month after launch.
39% OF OPERATORS
run on a custom domain
Pointed at real customers.
33% OF DEPLOYED APPS
kept building after launch
Came back to keep building a month after launch.
39% OF OPERATORS
A prototype rarely earns its own domain. Neither does it get a second month of development.
The payoff
For one in three operators, there was nothing to replace,
because nothing had been possible before. It exists because they built it, or it does not exist at all. For the rest, it replaced a cost they were already carrying:
The biggest unlock, by operator
Source: 300 structured operator interviews
Operator stories
The businesses behind the numbers
Nine operators, named with consent, on what broke, what they built, and what it changed. Figures in these stories are as reported by the operators themselves.
Drag to explore. Click a card to read the full story.
Closing
What this index observed
The software running these 50,000+ apps was not commissioned, purchased, or outsourced. It was built by the people who needed it, in the shape their business actually works, most of it by operators who had never built software before. For one in three of them, it could not have existed at any price they could pay. Now it takes bookings, moves inventory, collects rent, and sends receipts, every day, for businesses that stopped waiting for someone else to build it.
Sources: Emergent platform data (deployed-app cohort) and 300 structured operator interviews. Prompts are operator-verbatim, lightly cleaned for filler, and anonymized by business. Exact bases, query-level detail, and the measurement basis for end-user figures live in the methodology appendix.
Appendix
Methodology
How every figure in this report is measured, and on what base.
THE UNIVERSE
This report studies businesses on Emergent with at least one deployed, live app: a sample set of 50,000+ apps as of the report snapshot. Deployment is the inclusion bar throughout. Apps that were built but never deployed are excluded from every figure.
PLATFORM FIGURES
All behavioral figures (custom domains, payments, transactional email, integrations, return-to-build) come from Emergent platform data. Each figure is computed on its own measurement base rather than forced onto a single denominator. Custom domain, payment, and transactional email shares are computed on the deployed-app universe. Integration counts are computed per app across apps built on the platform, excluding the default database that ships with every app; counting it would inflate every figure. The full distribution: ~59% added at least one integration, ~32% two or more, ~16% three or more, ~6% four or more. Return-to-build (39%) is computed per operator: the share still actively building on an app one month after its launch. The real estate use-case share (about a quarter of builds) is computed from platform build data, not from interviews or prompts.
END USERS
The eight million figure counts people using the sampled deployed apps: customers, clients, and staff on the other side of the software. It is measured on this sample only and is a floor, not a platform total.
OPERATOR INTERVIEWS
300 structured interviews with operators from the deployed cohort. Interview findings describe this group and are not extrapolated to the full population. The coping distribution (33% / 27% / 27% / 12%) is computed among the operators who were not already well served by existing software, roughly six in seven of those interviewed. Development cost figures are quotes these operators reported receiving; $20,000 is the median, not the mean.
PROMPTS
The prompts in the industry portraits are operator-verbatim, lightly cleaned for filler, and anonymized by business. They illustrate language and range within a vertical. No proportion claim in this report is drawn from prompts.
OPERATOR STORIES
Published with operator consent. Figures in stories are as reported by the operators themselves and are attributed to them, not to platform data.
ROUNDING
All figures are rounded. Shares are decimal-free by design; precision beyond a whole percentage point would overstate the resolution of the underlying data.
Further reading
Relevant posts & resources
"Emergent Is Now a Unicorn: We Raised $130M in Our Series C at a $1.5B Valuation", Emergent News
"How a Global Industrial Equipment Manufacturer Replaced a Six-Figure CRM Contract with a Tool They Built Themselves", Case Study
"How the CEO of an Event Agency Built His Own Operations Platform for €1,000", Case Study
"Construction Contractor Built a Crew Dispatch and Scheduling Tool to Coordinate Jobsite Assignments", Case Study