InstaLILY announced a $60 million Series B on July 14, 2026, led by Energize Capital with participation from Home Depot Ventures, United Rentals, and Insight Partners. The company has raised nearly $100 million total. Revenue grew five times over the past year. The product is called Lily, and it does something most AI companies refuse to talk about: it stays.

Lily is what InstaLILY calls an AI forward deployed engineer. It goes into a business, learns how the work actually happens, builds custom software around those workflows, and ships it live in days. Not weeks. Not quarters. Days. Then it keeps running, adjusting, and rebuilding as the business changes underneath it.

That last sentence is the entire pitch. And it is the part that should concern you if your organization bought AI and then watched it slowly stop working.

The Decay Problem

Most AI deployments follow the same arc. An engineer or consultant builds the system. It works in the demo. Leadership signs off. The builder moves on to the next project. And the system starts drifting the day they leave.

Data formats change. Business rules shift. New tools get added. The model that powered the original workflow gets updated or deprecated. Nobody on the team knows how the system was wired together. Six months later, it still technically runs, but it produces answers nobody trusts, so the team routes around it and does the work manually.

McKinsey has documented this pattern repeatedly. Forty-three percent of AI initiatives fail to move past pilot. Gartner reported in early 2026 that 85 percent of enterprises pursuing agentic AI are not ready for production deployment. Those numbers do not describe a technology failure. They describe an operations failure. The AI worked. The organization did not have the layer to keep it working.

InstaLILY built that layer and turned it into a product.

What the Numbers Actually Show

The company’s customer results tell you more than the funding round does.

At a national distributor, Lily built software that finds and pursues revenue across the entire customer base. The result: more than $200 million in new annual sales identified, beyond what any human sales team could cover by hand. This is not a chatbot answering questions. This is software that was built by an AI, deployed to production, and generating measurable revenue.

At a field service company, a technician used to spend 15 minutes diagnosing each problem before starting the repair. Lily cut that to under 10 seconds. The cost to serve a single call dropped 98 percent.

A logistics operator that planned routes in 15 minutes now does it in three. A services company cut the time to train a new field rep by more than 60 percent.

None of these results came from a better model. They came from software that was built around the specific way each business operates, deployed fast, and kept running.

The Strategic Investors Tell the Real Story

When Home Depot Ventures and United Rentals invest in an AI company whose product they also use, that tells you something the press release does not. These are not financial bets. They are operational bets. The corporate investors want first-look access and pricing leverage on a tool they have already seen work inside their own operations.

Patrick Garcia, the Chief Digital and AI Officer at SRS Distribution (part of the Home Depot family), said it directly: “Lily has helped us build the software our teams need while working within the enterprise platforms, governance, and processes we’ve already established.”

That sentence describes the opposite of what most AI adoption looks like. Most organizations buy a model, run a pilot in a sandbox, and then spend six months trying to connect it to the systems their team actually uses. SRS Distribution embedded an AI that builds directly inside the existing infrastructure. No sandbox. No six-month integration project.

Why This Matters for Your Organization

You do not need to buy InstaLILY’s product to take the lesson.

The lesson is that the hardest part of AI is not selecting the right model, passing the proof of concept, or getting budget approval. The hardest part is the ongoing operational work of keeping AI connected to how your business actually runs, adapting it when the business changes, and making sure someone owns the system after the initial build.

Most organizations treat deployment as a one-time event. Build it, ship it, move on. That works for traditional software because traditional software does not drift. You install a CRM and it works the same way a year later. AI does not work that way. The data changes. The models change. The business context changes. Without an operations layer that absorbs those changes, every AI deployment has an expiration date.

InstaLILY’s $60 million bet is that the market for that operations layer is massive, because almost nobody has built it. They are probably right. BCG estimates that roughly 5 percent of companies generate outsized value from their AI investments. Close to 60 percent report little material return despite real spending. The gap between those two groups is not the model. It is whether someone stayed after the install.

The Question You Should Be Asking

If you run a team that uses AI in any form, ask this: who owns the system six months from now? Not who built it. Not who approved the budget. Who is responsible for the ongoing work of keeping it connected, accurate, and useful as your business changes around it?

If the answer is nobody, you already know what happens next. The system drifts. The team loses trust. The work goes back to manual. And six months after that, someone in leadership asks why the AI investment did not produce results.

The investment produced exactly what it was set up to produce. A demo. The part that produces results is the part that comes after, and it is the part that $60 million just went to fund.