I originally wrote “Evaluating Technology for Opportunity” for Managing Technology for Profit: A Small Business Guide as a practical framework for identifying, evaluating and acquiring technology-based business opportunities.
The original approach reduced technology due diligence to three fundamentals—People, Protection and Price—using common sense to bridge the gap between technical innovation and commercial enterprise.
This 2026 edition revisits that framework for a world transformed by AI, software, data and dramatically shorter technology cycles. The fundamentals remain remarkably durable, but the questions have changed. AI can accelerate development, lower barriers to entry and rapidly commoditize what once appeared proprietary.
For that reason, I have retained the original three Ps and added a fourth—Pace—to reflect the increasing importance of how quickly an opportunity can be developed, defended and commercialized before the technology or market moves on.
EVALUATING TECHNOLOGY FOR OPPORTUNITY
2026 Edition
Over the years, I developed a number of surveys for screening technology-based business opportunities. In venture capital parlance, this is called due diligence.
The purpose of due diligence is not to prove that an opportunity is good. It is to discover why it might be bad before we put our money into it.
The basic questions have changed surprisingly little:
Is the technology technically feasible?
Is the product a breakthrough or merely an improvement on what already exists—what I used to call “cutting the corner” on technology?
If it is an improvement, what significant advantage does it offer? Is the advantage obvious to the customer?
What does it cost to make and deliver?
Is there a sufficient market? Does that market already exist, or do we have to create it?
Who is going to buy it—and why?
How developed is the technology?
How much money and time are required before we have a product customers will actually pay for?
How large is the window of opportunity?
What protects us?
Who owns the technology and the associated intellectual property?
Who are the major players?
What competing technologies can solve the same problem?
And in 2026 we need to add some new questions:
What happens if AI makes this technology dramatically cheaper or easier to reproduce?
Could an AI-enabled competitor enter the market faster than we can establish it?
Does the opportunity depend on somebody else’s AI model, cloud platform, software, data or intellectual property?
Do we own—or have secure rights to—the data necessary to operate the business?
Can the technology scale economically and securely?
What regulatory, privacy, cybersecurity or AI-governance issues could restrict commercialization?
What happens when today’s technology becomes tomorrow’s commodity?
In marketing, the traditional four Ps—product, price, place and promotion—help match a product or service to its customer and market.
Taking a lead from this elegant approach, I developed a simpler three-P system for evaluating technology opportunities:
People. Protection. Price.
It was designed to use basic people skills and common sense to build a decision bridge between high-powered science and enterprise.
In 2026 those three Ps remain fundamental.
But I would add one more:
Pace.
PEOPLE
With desktop publishing and high-priced consultants, high-tech proposals once could be made to look like winners.
AI has taken this to another level.
Today almost anybody can produce a polished business plan, impressive presentation, market analysis, financial forecast, product demonstration or technical report.
Consequently, polish tells us less than ever about the underlying opportunity.
Business remains, first and foremost, a people science.
Meet the principals.
Ask the inventor, scientist or entrepreneur to explain the opportunity without the slides.
Start with the original question:
In one simple sentence, what is the opportunity?
If the answer requires ten minutes of technical explanation, we may have a technology looking for a market rather than a business opportunity.
Then ask:
What problem does it solve?
Who has that problem?
Who will pay to solve it?
Why is this solution better than what the customer uses today?
What evidence do you have that customers care?
These questions test more than communication ability. They force the technology into a commercial context.
Then qualify the people.
What role does the inventor want?
Does the entrepreneur want to build the business or simply find somebody else to finance and develop the idea?
Who else is needed on the team?
What expertise is missing?
How do the principals respond when their assumptions are challenged?
Ask one particularly useful question:
What could cause this opportunity to fail?
An entrepreneur who understands the weaknesses in an opportunity is generally more useful than one who believes there are none.
Look for enthusiasm, commitment and staying power—but also intellectual honesty and adaptability.
Technology ventures rarely develop according to the original plan.
The people we back therefore need the ability to learn, change direction and keep going when the original assumptions prove wrong.
AI makes technical capability easier to acquire.
It does not manufacture judgement, integrity, leadership or staying power.
Qualify the people before falling in love with the technology.
PROTECTION
If the people pass the hurdle, take a serious look at how the business will survive success.
There is little future in blazing a trail if the latecomer can capture the market after we have spent the money developing and educating it.
In crass terms, technology-based businesses still sell two valuable commodities:
time and protection.
Time to leapfrog the competition.
Protection to establish an embryonic business beyond the competitor’s immediate reach.
Traditionally, the primary protection was the patent.
Patents remain important. But the key question was never simply whether a patent existed.
It was:
How strong is the position?
What do the patent claims actually cover?
Can competitors design around them?
Where have applications been filed?
Who developed the invention?
Who paid for the R&D?
Who owns the rights?
Have any rights been licensed, assigned, pledged or otherwise encumbered?
Has anything been publicly disclosed that affects protection?
And importantly:
Do we have freedom to operate?
Owning a patent does not necessarily mean we can commercialize a product without infringing somebody else’s rights.
In 2026, protection must be viewed more broadly.
It may come from:
patents, trade secrets, copyright, trademarks, exclusive licences, proprietary software, proprietary data, regulatory approvals, manufacturing know-how, distribution rights, customer relationships, network effects, switching costs or brand.
It may also come from speed.
A business that continuously learns and improves faster than competitors can sometimes create a more formidable barrier than a single patent.
AI makes this especially important.
Ask:
Could a competitor reproduce the important part of this technology using a generally available AI model?
Are we building proprietary capability—or simply wrapping somebody else’s technology?
What happens if the AI model we depend upon becomes freely available?
What happens if the supplier changes its price, licence terms or access?
Does customer use generate proprietary data or know-how that strengthens our position over time?
The strongest opportunities increasingly have layers of protection.
A patent may protect the invention.
Trade secrets protect the process.
Data improve the product.
Distribution provides customer access.
Brand creates trust.
Switching costs retain customers.
Continuous innovation keeps the company ahead.
A competitor then has several fences to cross rather than one.
The question is no longer simply:
“Is it patented?”
The better question is:
“If this succeeds, what prevents somebody else from taking the market we create?”
If the answer is merely “our technology is better,” keep asking.
THE AI TEST
AI deserves a specific test because it can be both the opportunity and the destroyer of the opportunity.
Ask four questions.
1. IS AI ESSENTIAL?
Does AI materially improve the economics, performance or customer value of the product?
Or has AI simply been added because the market currently rewards the label?
Technology looking for a problem is still technology looking for a problem.
2. WHAT DO WE ACTUALLY OWN?
Is there proprietary technology?
Proprietary data?
Specialized know-how?
A protected workflow?
Customer relationships?
Or are we simply using the same models and tools available to everybody else?
There is nothing inherently wrong with building on somebody else’s platform.
But do not mistake access for ownership.
3. WHAT COMPOUNDS?
A good technology business should become stronger as it grows.
Does each customer create additional data, knowledge, distribution, reputation, network effects or recurring revenue?
Does experience improve the product?
Does the company become harder to displace?
If nothing compounds, competitors may eventually compete the advantage away.
4. WHAT HAPPENS WHEN AI GETS BETTER AND CHEAPER?
Assume that today’s expensive capability eventually becomes inexpensive.
Then ask:
What is left of our competitive advantage?
If the answer is “not much,” we may not own a technology business.
We may simply be renting a temporary technology gap.
PRICE
Price is the next hurdle.
The original principle remains sound: the selling price must support the costs of developing, producing, distributing and servicing the product while providing an adequate return for the capital and risk involved.
But technology economics have changed.
For a manufactured product, unit production cost may remain critical.
For software and AI, the marginal cost of another customer can be low while the costs of developing the product and acquiring that customer can be substantial.
Calculate the whole economic system.
Include:
- research and development;
- manufacturing;
- software development;
- cloud and computing costs;
- AI-model usage;
- data acquisition and licensing;
- cybersecurity;
- regulatory compliance;
- implementation;
- sales and marketing;
- customer acquisition;
- distribution;
- technical support; and
- continuing product development.
Then turn the calculation around.
Do not begin with:
What does it cost us?
Begin with:
What is it worth to the customer?
Does the technology save labour?
Reduce downtime?
Increase production?
Improve quality?
Accelerate decisions?
Reduce risk?
Increase revenue?
Replace an existing expense?
If we can quantify the customer’s economic benefit, we have the beginning of a rational pricing strategy.
The original rule of thumb that a high-tech product might require a substantial markup over production cost was appropriate to physical products moving through multi-tier distribution.
It is no longer universally appropriate.
A software product costing very little to reproduce may create enormous customer value.
A sophisticated manufactured product may have high production costs and relatively narrow margins.
The better rule for 2026 is:
Understand the customer’s economics before setting your own price.
Do not compete with a major competitor simply by cutting prices.
That is why we sought a protected niche in the first place.
Compete on value.
PACE
The fourth P is Pace.
The original technology evaluation process asked:
How large is the window of opportunity?
That question has become critical enough to deserve its own category.
AI, cloud computing, global communications and readily available development tools have compressed technology cycles.
Ask:
How quickly can we develop the product?
How quickly can we prove customer demand?
How quickly can we obtain regulatory approval?
How quickly can production scale?
How quickly can competitors respond?
How quickly will the underlying technology improve?
How quickly will our cash disappear?
And perhaps most importantly:
How quickly can we learn?
Technology once classified as seven to ten years from market, three to seven years from market and fully developed often followed relatively understandable development paths.
Today’s development paths can be less predictable.
A technical barrier expected to take five years to overcome may disappear when another company releases a new model, semiconductor, software library or manufacturing process.
Conversely, a product that can be developed in six months may still take years to achieve regulatory approval or customer adoption.
Therefore distinguish between:
technology pace, market pace and business pace.
They are not necessarily the same.
Being technologically ready before the customer is ready consumes capital.
Entering after the market has moved consumes opportunity.
Timing remains the entrepreneur’s art.
PUTTING THE FOUR Ps TO WORK
The four Ps can be reduced to four questions.
PEOPLE
Are these the people we want to back when the original plan goes wrong?
PROTECTION
If the opportunity succeeds, what prevents competitors from taking the market we create?
PRICE
Can customers receive compelling value while we earn an attractive return?
PACE
Can we establish the business before the technology, competition or market moves past us?
If an opportunity fails one test, do not automatically discard it.
Find out whether the weakness can be fixed.
Weak people may require a stronger management team.
Weak protection may be strengthened through licensing, exclusivity, data or distribution.
Weak economics may improve through redesign or a different market.
Poor timing may simply mean waiting.
But know the weakness before investing.
THE GOLDEN RULE
If the opportunity still feels good, seek independent technical, intellectual-property and commercial advice.
Talk to customers.
Talk to people who understand the technology.
Understand the ownership.
Check the competition.
Test the economics.
Find the assumptions on which the business depends.
Then try to prove those assumptions wrong.
The golden rule in the evaluation game remains unchanged:
If we can’t evaluate it adequately, it’s not for us.
But there is an important qualification for 2026.
Do not confuse careful evaluation with slow evaluation.
Technology moves too quickly.
The old approach of allowing an evaluation process to run for eighteen months may provide time for the principals to develop trust—but it may also provide enough time for the opportunity to disappear.
Stage the commitment.
Test the important assumptions first.
Invest against milestones.
Learn before spending heavily.
Preserve the ability to walk away.
Technology opportunities remain partnerships.
Get to know the people. Understand what each party contributes and expects to receive. Make certain interests remain aligned when things go wrong as well as when they go right.
The original metaphor was to imagine the technology as a bride with a dowry of products and patents, and to court the bride with money, manufacturing, management, marketing and distribution.
The underlying principle remains useful, even if the language has aged:
Technology brings something to the partnership, and so must we.
Today the dowry may include patents, software, data, algorithms, trade secrets, regulatory approvals, customers and know-how.
We may bring capital, manufacturing, management, marketing, distribution and market access.
Make certain the assets are real.
Make certain the person offering them actually owns them.
Understand what they are worth.
Then determine whether combining them with what we bring creates something more valuable than either party could create alone.
Court carefully. Learn quickly. Commit progressively.
Technology changes.
Markets change.
AI accelerates both.
But the fundamentals remain remarkably durable:
Back good people. Protect what matters. Price for value. Move while the window is open.