Payments is the best example I’ve seen of how the “good enough” mentality prevents technology adoption. In the developing world, mobile payments were adopted at a rate that far outpaced that of developed, card-swiping economies. 

The inverse relationship is so sharp you might not believe it:

Credit to NFX for their original Leapfrog post.

NFX calls this the Leapfrog Effect: when an industry or market skips a step along the technology transformation chain. 

The argument goes: rather than evolve incrementally from analog to digital, a leapfrogging company goes from digital zero straight to tech-first hero. Usually, this occurs for one reason: there is only upside, no risk. Consumers in emerging markets did not develop the habit of using credit cards; businesses were not built around them; and brand trust in credit card companies didn’t exist. So, when mobile payments came along, consumers easily adopted it because they did not have an existing, satisfactory solution. In the US, we have the opposite situation: we have infrastructure, cultural expectation, and trust in using credit cards for payments. Why would we adopt mobile payments when credit cards work just fine?

AI-adoption presents a similar situation for companies. For those that have sat on the Saas sidelines, now is the opportune time to reconsider core tooling. And much like how phones enabled frictionless digital payments, the general availability of AI models today - both paid and free - offer relatively frictionless access to sophisticated technology previously only available to “tech” companies. 

So who leapfrogs?

If we assume the plot above translates well to the pen-and-paper-to-AI-transformation space, we replace “credit card adoption %,” with “embedded Saas %.” Industries like finance, retail, ecommerce, social media, streaming, edtech, and others, can more easily get caught in AI-pilot-hell because the new systems must prove outsized value over the old ones. In two words, “good enough” can kill advancement in companies that have achieved modest tech adoption.

Conversely, if your company doesn’t feel “good enough” about your existing set up, you’re a candidate to leapfrog. You maybe…

• Cannot scale in volume because you’re restricted by slow, manual processes

• Don’t deeply understand what drives cost and revenue because you’ve never gotten forensic with your data

• Haven’t been able to justify a full-time hire for data, automation, or AI (let alone afford the accompanying tech stack)

• Don’t know what you don’t know: You’re looking at your business and wondering if AI can help in ways you can’t imagine

These scenarios precede outsized impact from strategic integration of data, automation, and AI. Let’s look at an exemplar from the trades.

Let's take a look at an exemplar from the trades

View the original Instagram post

Tommy Mello is the founder of A1 Garage Doors, a garage door install, repair, and service company with more than $300M in annual revenue. Mello undoubtedly does a lot of things right. But in listening to him on a recent podcast - you realize he is a leapfrogger in his industry when it comes to data and AI:

1. Operational analytics: 

“... I know every single one of my CSR's [customer service representative’s] booking rates. I know exactly what we're training them on every day. I know exactly my technique … You should see my scorecards. I mean, these are world-renowned … they are the baddest ass thing you'll ever see in your life … The data, the KPIs will set you free.”

2. Experimentation:

“...yesterday, my AI agents were [at] 87% booking rate. My real agents were at 92%. So it's a 5% variance. But the AI continues to get better.”

3. Machine learning and prediction:

“... what our dispatch software does is regression testing. So you could feed it data, credit card score, how much should they pay off in the home? How many garage doors do they have? And then it looks for outliers. And it knows what technicians do well on this type of job.

4. North-star metrics:

“... And the most important thing is customer satisfaction. Net promoter score and my internal net promoter score. So I'm not just looking to sell.”

You can plainly see that Mello doesn’t just use general software; he leapt to the upper echelons of data strategy. From the sales motion, to algorithmically matching customers and technicians, to retention and net promoter scores - data is a driving force throughout A1 Garage Doors. 

Each of these four examples were possible before artificial intelligence was mainstream. But they required really significant tech, talent, and time investments – things generally not available to non-tech companies.

Tying it all together

LLMs are bringing down the barrier to entry for leveraging sophisticated technology. Those that were previously not engaged with tech are starting to swing from pens to LLMs. To be clear, that severe whiplash can be painful (it’s culturally very challenging to go from data-zero to AI-enabled, but I’ll leave that for a different blog.). 

However, there has never been a more inviting time to hop aboard the data train. 

All it takes is some strategy, a little bravery, and a willingness to resist the urge to stop at “good enough.”