AI in a 130-Year-Old Company: How Heritage Businesses Should Adopt New Technology Without Losing Their Soul

Our company was founded in 1890. The first Manzanos vineyards were worked by hand, the ledgers were kept on paper, and the only "network" was the relationships our family built one harvest at a time. Today we run eight active business verticals across more than 75 countries, and on my desk this morning there was a model output forecasting demand for a wine that will not be bottled for two years.

That gap — between a hand-pruned vine and a demand model — is where most heritage companies get technology badly wrong. And they get it wrong in two opposite directions at once.

Some treat every new tool as a threat to the craft and adopt nothing, until a younger, faster competitor quietly eats their margins. Others, terrified of being left behind, adopt everything — they buy the software, hire the consultants, announce the "digital transformation," and three years later have a more expensive version of the same business, plus a graveyard of abandoned platforms. I have watched both failures up close. Neither is necessary.

## The two ways heritage companies fail at technology

The first failure is **romantic paralysis**. It sounds noble: "We do things the traditional way." In wine, in hospitality, in real estate, there is real value in the human craft — and that becomes the excuse to refuse every tool, including the ones that have nothing to do with craft at all. Refusing to automate your invoicing because you respect the winemaker's hands is not tradition. It is just inefficiency wearing tradition's clothes.

The second failure is **transformation theater**. A board gets nervous, a buzzword arrives — three years ago it was "blockchain," today it is "AI" — and the company launches an initiative measured by activity rather than outcome. Money is spent, decks are presented, and nobody can point to a single decision that got better or a single customer who is happier.

Both failures share a root cause: the company never decided what it was actually trying to protect and what it was trying to fix. Technology adoption is not a values question. It is a clarity question.

## The rule we use: automate the friction, never the soul

At Manzanos Enterprises we run every technology decision through one distinction. Every business has a **soul** — the small number of things that are the reason customers choose us and that must stay human — and it has **friction**, the unglamorous work that consumes time, introduces errors, and creates no loyalty whatsoever.

- The soul of a winery is the blend, the decision of when to pick, the judgment in the cellar, the relationship with a distributor who trusts your word. Nobody buys a Rioja because your accounts reconciled faster.

- The friction is reconciling those accounts, chasing invoices, re-keying orders between systems, drafting the tenth version of a routine contract, forecasting how much glass to buy, answering the same booking question for the hundredth time.

The rule writes itself: **automate the friction, never the soul.** A vineyard manager should spend more hours among the vines and fewer hours in spreadsheets — so we put the tools on the spreadsheets, not the vines.

Concretely, here is where new technology, including AI, earns its place in a heritage business:

- **Demand and inventory forecasting.** A model that predicts how a wine will sell across markets lets us buy dry goods and plan production with less waste and less tied-up cash. The winemaker still decides what to make; the model decides how much glass to order.

- **The first draft of everything administrative.** Routine contracts, supplier correspondence, translations across the languages we operate in, internal reports. AI produces the draft in minutes; a human edits and owns it. The leverage is enormous and the risk is contained, because nothing goes out the door unread.

- **Customer service on the repetitive 80%.** In hospitality at Palacio de Manzanos and in our residential business, most inbound questions are the same dozen questions. Handling those instantly frees our people for the conversations that actually require a human — the ones that close a sale or save a relationship.

- **Pattern-finding in data we already own.** Decades of sales, harvest, and cost data contain answers we never had time to extract by hand. Tools surface the pattern; experienced people decide whether it means anything.

## Three tests before we adopt anything

Enthusiasm is not a strategy, and neither is fear. Before any tool enters the business, it has to pass three tests.

**1. Does it move a number that matters?** Not "is it innovative" — does it reduce cost, increase revenue, cut errors, or free up senior time we can redeploy? If we cannot name the metric before we start, we are buying theater. We pilot in one vertical, measure honestly against that number, and only then roll out. Most pilots do not graduate, and that is the system working.

**2. Does it keep a human accountable?** We do not let a model make a decision that no person is willing to sign their name to. AI drafts the contract; a person signs it. The model flags the credit risk; a person grants the credit. This is not nostalgia — it is risk management. The moment nobody owns the output, you have built a liability, not a capability.

**3. Will it still be ours in five years?** A 130-year-old company thinks in decades, and that discipline applies to technology too. We are wary of tools that hold our data hostage, that we cannot operate without a permanent consultant, or that a vendor can switch off or reprice at will. We would rather adopt a simpler tool we control than a sophisticated one that controls us.

## What the craft industries got right all along

There is a quiet irony in watching the technology world rediscover ideas that heritage businesses have lived by for over a century. The most durable companies have always understood that scarcity, provenance, and human judgment carry a premium — and as machines do more of the routine work, that premium grows, it does not shrink.

The more automated the world becomes, the more people pay for the handmade, the authentic, the human. A bottle of wine made by a person who decided, in a specific September, that the grapes were ready — that is worth more in an AI-saturated world, not less. The same is true for a hotel stay where a real person remembers your name, or a building designed by someone who walked the land. Technology should protect the time and economics that make that human craft possible. It should never replace the craft itself, because the craft is the product.

This is why I am genuinely optimistic about AI in traditional industries — and deeply skeptical of how most of them are approaching it. The opportunity is not to become a technology company. It is to use technology to become a more focused version of the company you already are: less time on friction, more time on the work only you can do.

## Key Takeaways

- Heritage companies fail at technology in two opposite ways — adopting nothing out of romantic paralysis, or adopting everything as transformation theater. Both come from not deciding what to protect and what to fix.

- Separate your business's **soul** (the human judgment customers actually pay for) from its **friction** (unglamorous work that creates no loyalty). Automate the friction relentlessly; never automate the soul.

- The highest-return uses of AI in traditional businesses are forecasting, first drafts of administrative work, repetitive customer service, and finding patterns in data you already own.

- Run every tool through three tests: does it move a number that matters, does it keep a human accountable, and will you still control it in five years.

- Pilot in one unit, measure against a named metric, and let most pilots fail — that is the system working, not breaking.

- As automation spreads, the premium on the handmade and the human rises. Technology should buy back the time that makes craft possible, not replace the craft.

- A company that thinks in decades should adopt technology the same way it adopts everything else: deliberately, on its own terms, and never at the cost of what made it worth preserving.

We did not survive 130 years by resisting change, and we will not survive the next 130 by chasing it. The companies that endure are the ones that know the difference between the tools and the trade — and never confuse the two.

*Meta description: Heritage businesses fail at technology by adopting nothing or everything. How a company founded in 1890 thinks about AI — automate the friction, never the soul.*

*SEO keywords: AI in traditional industries, technology adoption heritage business, digital transformation family business, automate not replace, AI risk management, innovation without losing soul, long-term technology strategy, AI in wine and hospitality*