In the new issue of Harvard Business Review, Wharton Professors Nicola Siggelkow and Christian Terwiesch propose new data-driven business models that involve developing deeper, longer-lasting, and more valuable relationships with more customers by making strategic use of today’s continuous flow of data on customers.
Profs. Siggelkow and Terwiesch start from the premise that most current businesses operate using a buy what we have business model, meaning businesses wait until customers identify what they need and then come to the business to buy what the business has (thus, buy what we have).
The need to change…
The need for change is rooted in the competitive race to provide customers with better, more customized product/service offerings with the customer having to expend the least amount of time and energy.
The authors argue that with all the data being continuously and freely generated as people use their computers, phones, apps and smart devices to interact with businesses, businesses must continuously and increasingly exploit this data if they want to survive in a competitive marketplace and then to thrive.
The reason that adopting business models that rely on continuously exploiting data is so essential to competition and survival is the powerful feedback loops involved. That is, “[t]he better the company understands a customer, the more it can customize its offerings to her. The more delighted she is by this, the more likely she is to return to the company again, thus providing it with even more data. The more data the company has, the better it can customize its offerings. Likewise, the more new customers a company attracts through its superior customization, the better its population-level data is. The better its population data, the more it can create desirable products. The more desirable its products, the more it can attract new customers. And so on. Both learning loops build on themselves, allowing companies to keep expanding their competitive advantage. Over time these two loops have another very important effect: They allow companies to address more-fundamental customer needs and desires.”
What to do?
The authors posit that businesses should continually gather and use these large volumes of customer data in order to more proactively (and helpfully!) interject themselves into all four stages of the traditional customer journey so that the customer has to expend increasingly less time and energy meeting their needs. The four stages that business need to more efficiently and effectively exploit are:
1. The Recognition stage: where a customer becomes aware of a need
2. The Request stage: where a customer identifies a product or service that would satisfy this need and reaches out to a company
3. The Respond stage: where a customer experiences how the company delivers the product or service.
4. The Repeat stage: where the customer plans for and executes on ongoing needs
This is where the authors actually make a contribution to business theory: The authors propose four business models that operate to continuously inject the business deeper into the stages of the customer journey. Note that since companies are likely to have customers with different preferences, most firms will have to create a portfolio of several of these connected strategies/business models.
Business Model 1: Respond to Desire
Businesses that deal with customers who know exactly what they want, and who want their needs met on demand should adopt the Respond to Desire (R2D) business model. This business model requires optimizing the customer experience in the Respond stage of the customer journey.
The R2D business model makes use of continuous, vast stream of customer/market data to optimize inventory and logistics so that they can provide fast delivery, minimal friction, flexibility and precise execution for each customer and for each transaction. Additionally, these businesses should make the execution of the transaction as effortless for the customer as possible.
After the transaction with the customer is completed, Businesses using the R2D business model should use the data on the transaction to make repeating the transaction as easy as possible (for example, using saved product/service preferences and payment/shipping information or making inventory decisions that would make future transactions faster).
In addition, R2D businesses should use the volume of transaction information at the population level to make product/inventory decisions and invest in scaling the most popular product/services so that transactions become faster and more effortless (e.g. Amazon’s scaling their inventory and logistics capabilities wherever the data points).
Business Model 2: Curated Offering
Businesses that deal with customers who need to select products from among an overwhelming number of options and who are willing to share information should develop the Curated Offering (CO) business model. This business model requires optimizing the customer experience in the Request stage of the customer journey.
The CO business model makes use of continuous, vast stream of market/customer data to make useful and meaningful product/service suggestions and recommendations. These businesses should make use of available customer data to ensure that the recommendations as relevant to each customer as possible.
After the transaction with the customer is completed, Businesses using the CO business model should use the data on the transaction to present even more useful recommendations in more areas of the customer’s needs than what the customer would otherwise be exposed to or would choose from amongst on their own.
At the population level, companies should use learnings from the continuous flow of data to invest in more relevant portfolios of product/service offerings and capitalize on potential packages of popular products/service offerings.
Business Model 3: Coach Behavior
Businesses that deal with customers who could benefit from overcoming inertia and biases in their product and service decisions/investments and who are willing to share personal data and get suggestions should develop the Coach Behavior (CB) business model. This business model requires optimizing the customer experience in the Recognition stage of the customer journey.
The CB business model makes use of continuous, vast stream of market/customer data to predict how and when a customer could benefit from a product and service offering. These businesses should make use of available customer data to track when the timing and product/service offering is advantageous to the customer and then to effectively communicate those suggestions to the customer
After the transaction with the customer is completed, Businesses using the CB business model should use the data on the transaction improve the timing, usefulness, and relevance of the suggestions.
At the population level, companies should use learnings from the continuous flow of data to invest in ensuring that customers continue to realize gains from suggested products services so that customers share more information and deepen the advisory relationship.
Business Model 4: Automated Execution
Businesses that deal with customers who have very predictable behavior (and for whom costs of mistakes are small) and who are comfortable partnering with other businesses should develop the Automated Execution (AB) business model. This business model requires optimizing the entire customer experience.
The AB business model requires companies to be awarded the full trust of their customers so that the customer is willing to share all relevant data with the company and to entrust the company with decision/purchasing authority on the customers’ behalf. These businesses should make use of this customer data to determine the best timing and product/services for a customer’s circumstances and then to execute on these determinations to the customer’s benefit.
Though this is a continuously replenishing process, the business using the AB business model should be learning from each customer transactions so that future replenishments and offerings are more helpful to the customer and that any slight errors are eliminated.
These learnings should be developed at the population level as well so that investments can be made in improving operational execution and scaling product/service offerings.
Continuously Repeat and Invest
All of these strategies require continuous, real-time learning from the continuous flow of data (most likely using machine-learning/artificial intelligence capabilities). The conclusions drawn from the data as to the most advantageous (from the customer and company perspective) product/service offerings can be used to maximize returns and then to reinvest in emerging product/service offerings and capabilities.