Author: William Vorhies
Summary: McKinsey says platform companies will represent 30% of global business revenue by next year (2020). Here are some lessons and examples to help mature companies evaluate where they can create AI/ML-enabled platforms to remain competitive. This is a long topic so this will be Part 1 of 2.
As data scientists increasingly become critical resources in enabling companies’ exploration and exploitation of their digital resources, it’s also increasingly important that data scientists can provide accurate and focused business guidance. If that seems like a mouthful, try these two scenarios.
Scenario 1: Joe is asked to recommend a portfolio of AI/ML projects that will improve performance, provide measurable ROI, and have relatively low risk of failure.
Scenario 2: Joan is asked to plan AI/ML projects that will maximize the value of the company, protect against competitors, and create the fastest possible market, revenue, and margin growth.
Pretty much any of us could do a competent job with the first scenario. We’d look at the current business and try to find opportunities where traditional ML could be used like scoring, forecasting, or optimization. If we’re sufficiently advanced we’d also look for AI opportunities like the application of NLP or image processing. Since we weren’t asked to challenge the fundamental business model, we just looked to places where we could paste on AI/ML.
In our previous article we called this “AI-Inside” or “Applied AI”. It’s not necessarily bad to do this. It’s walk-before-you-run; get some experience; earn some buy-in. What it’s not in any sense of the word is transformational.
Scenario 2 invites Joan to look deep for opportunities to really change the business with AI/ML. As we identified in our last article, this transformational strategy is increasingly called ‘Platform Strategy’, or ‘Platformication’.
Here’s the kicker. McKinsey says that by 2020 (next year!) platform businesses will account for 30% of global business revenues. So if you’re just now finding out about platform strategy, you and your company are about to see the backsides of many of your competitors as they rush ahead to take your customers and disrupt your markets. It’s no exaggeration to say that platform strategy represents the basis for the next industrial revolution in worldwide business.
So the first part of this lesson is this: if you’re asked to help your company define its ‘digital journey’ because your CXO read an article that said you have to get value from your data, don’t generalize. Be prepared to engage in a conversation about whether this is incremental change or whether AI/ML can create really wide-moat transformational change.
Platform Strategy
We devoted the last article to providing a definition of Platformication and some structure around how to think about it. If you haven’t read that, start there first.
Here are some bullet points to get you back into thinking about this:
- The value driver of this strategy is a network you create. The more users the more valuable your network becomes.
- The ‘platform’ itself is an intermediary electronic exchange that at its core is a matching system between ‘buyers’ and ‘sellers’.
- You won’t necessarily be focusing on your own products and services. They may be part of the offering but the network effect only grows if you let other sellers into the tent (think Amazon) which in turn attracts more buyers. Better to have 5% of a billion dollar market than 95% of a $10 million market.
- AI/ML is the secret sauce that allows you to make the users’ experience seamless, frictionless, efficient, pleasurable, and valuable. That’s going to be a combination of financial incentives like wide choice and best price, but also a heap of intangibles broadly defined as value.
- In fact, you can build a platform without any financial exchange at all simply by providing good content and becoming a trusted advisor, increasing the likelihood that the customer will look first at your offerings because of that built up good will and frequent visits to your site.
- Finally, the information and interactions you record on the platform provide the training data for improving the UX. There are also many opportunities to create an information product from that data which may be valuable to you in future planning, or can be sold to other suppliers in your ecosystem.
No Agreed Path to Platformication
We’ve grown comfortable and familiar with the famous tech names in Platformication that started out with this strategy as a goal. Amazon, eBay, Etsy, Airbnb, Uber, OpenTable, Travelocity all set out to create businesses in which they owned none of the delivery resources, had no inventory, and had only the goal of creating large ecosystems of buyers and sellers. As they say, these now-giants weren’t encumbered by their own prior success.
What’s important to our exploration here is how mature companies can identify platform opportunities and how those can be implemented alongside their existing business models or even completely replace them. We’ll start with some lessons learned and guidelines and then illustrate them with company examples.
Lessons, Guidelines, Examples
- Information centric businesses are obvious targets. (E.g. insurance, mortgage lending, media, telecom, real estate brokerage).
- Since network size is the measure of success, it is more likely these will be B2C.
MassMutual Insurance
MassMutual has embraced several flavors of platform. Its RetireSMART platform is an interactive content site where individuals planning for retirement can view quality content as well as several financial forecasting tools designed to show readiness for retirement.
MassMutual partnered with Envestnet – Yodlee data aggregation platform to bring together client assets from other financial firms including 401Ks that may still be housed at prior employers.
One lesson is that it’s not necessary to build all the features of the platform yourself and partnering may be an option as it was here.
- Fragmented industries are good targets.
Other Insurance Companies, Distributors, and Agents
Other insurance companies are increasingly willing to cross-sell competitors’ products to maintain and consolidate their client relationships. The traditional insurance business is an inefficient pipeline style model relying on carriers, distributors, and agents.
Where information is fragmented, that is sellers know a lot more than buyers, platforms aggregate the supply side adding transparency and the ability to compare offers making the consumer’s experience nearly frictionless.
Carriers, distributors and agents benefit from a low cost solution to support the consumer and gain substantially in their ability to analyze the resulting data from interactions to better design products and target appropriate audiences.
Westlaw Edge
Here’s a slightly different platform model. Westlaw had long sold legal research information to the highly fragmented law office market. Their platform strategy however is to provide that information on a subscription basis enhanced by AI search capability.
The network benefit is to relieve lawyers of the tedious work of establishing citations in support of their cases. The super-benefit of using AI/ML enabled search is to alert the attorney for law that has been indirectly overturned and even alert for “Overruling Risk,” meaning the case has likely been implicitly overturned by another case. The software can accurately tag implicitly overturned cases 80 to 85% of the time.
- Best of all find fragmented markets that are under served.
La Ruche Qui Dit Oui
This French startup seeks to become the platform between local food producers (those most likely to participate in farmers markets) and consumers who want quality and freshness, and also want to support local farm-to-table culture.
The name translates rather inelegantly in English to roughly ‘the beehive that says yes’, or perhaps more accurately many small buyers and sellers coming together to make this transaction easy. Essentially they have added a layer of ecommerce to farmers markets allowing:
- On line ordering without a middleman.
- Collect your order in person once a week from a local venue (farmers market).
- And of course, eat better with fresh local produce.
On the supply side they promise:
- Suppliers (growers / farmers) keep control of price and minimum order quantities for delivery.
- Change their offering at any time based on seasonal availability.
- Know how much they’re going to earn and bring only what’s pre-ordered so nothing goes to waste.
- Quick payments and simplified accounting.
InsureMyTrip, Squaremouth, TravelInsurance.com
These are all platforms that have emerged recently and are in head-to-head competition around the highly fragmented travel insurance business. These are fairly complex policies purchased by vacationers who want to protect their investment from unforeseen interruption. However, the differences among different policies (if it’s not written it’s not covered) and the fragmentation among relatively small providers results in a classic imbalance of knowledge between buyers and sellers.
Two lessons here:
- Yes there is competition among emerging platforms so first movers who execute well are favored.
- Any of these insurance carriers could themselves have been first movers but instead they’ve been outflanked by knowledgeable startups who reduce the carriers to commodity suppliers. Don’t wait to get started or you could end a commodity in someone else’s network platform.
This is a long topic with many lessons and examples. To keep this readable we’ll split this into two parts. Be sure to see the balance in our next article:
AI/ML Lessons for Creating a Platform Strategy – Part 2
Other articles on AI Strategy
A Radical AI Strategy – Platformication
Now that We’ve Got AI What do We do with It?
Capturing the Value of ML/AI – the Challenge of Offensive versus Defensive Data Strategies
The Case for Just Getting Your Feet Wet with AI
The Fourth Way to Practice Data Science – Purpose Built Analytic Modules
From Strategy to Implementation – Planning an AI-First Company
Comparing the Four Major AI Strategies
Comparing AI Strategies – Systems of Intelligence
Comparing AI Strategies – Vertical versus Horizontal.
What Makes a Successful AI Company – Data Dominance
AI Strategies – Incremental and Fundamental Improvements
Other articles by Bill Vorhies
About the author: Bill is Contributing Editor for Data Science Central. Bill is also President & Chief Data Scientist at Data-Magnum and has practiced as a data scientist since 2001. He can be reached at: