Software companies bundle support services with their products as standard practice. Is it possible to be different…and profitable?
The concept of bundling is not new—just think of the McDonald’s Happy Meal, which was first introduced in 1979. The benefits are well-articulated: increased efficiency, reduced marketing and distribution costs, and ultimately, increased profits.
Within the software industry, the practice of bundling is known as SaaS—software as a service. Salesforce is often identified as the company that sparked mass adoption of SaaS, making software available over the internet for a monthly fee that is now de rigeur for the industry. Adobe, the big daddy of graphic design, made the jump in 2012 when it stopped being a provider of one-time purchase products and became a SaaS provider with differently priced tiered options to cater to different customer needs.
While SaaS is the norm in the software industry, where technical support represent a major part of the service element, would the same model work for companies providing artificial intelligence (AI) solutions? What about unbundling the service aspect and then re-bundling it on demand? For Singapore-based company TAIGER and its founder Sinuhe Arroyo, that is the question.
A bundle of joy (and revenue)
Arroyo set up TAIGER in Singapore in 2015 after lukewarm response to his idea for a natural language processing (NLP) AI start-up in his native Europe. NLP is a sub-field of AI that deals with the interaction between computers and humans using natural language.
TAIGER’s NLP technology was adopted quickly in the country, partly due to the ecosystem and partly due to the solution’s readiness for the market. By 2020, the valuation of the company had crossed US$110 million. Unlike many other scientific inventions that focused on technology, the AI models offered by the firm were tailored to identify real business use cases like small and medium enterprise (SME) customers' on-boarding process or electronic upload process of customer forms.
Clear business use cases allowed easy identification and mapping of existing processes and provided scalable engineering options to meet customers’ business goals. The accuracy achieved by the solutions were high (about 95 percent) and were attributed to the efficacy of the technology and the discipline imposed by the strict service level agreements (SLAs) framed by the company.
The solutions offered by TAIGER were grouped under a unified platform featuring three bundled tools. Clients could purchase each of the bundled solutions individually or together under licensing contracts, which took care of customisation and implementation of the tools as per the client’s needs. Every license sold was treated as a project, and key performance indicators (KPIs) and SLA’s determined at the beginning of the project were used as monitoring guides to guarantee product accuracy.
Despite several successful implementations, there were several known challenges with the project-based approach. One consistent challenge was customisation, which often took up more time and effort than initially accounted for. When the client’s demanded more customisation, contractual liabilities enforced such requests resulting in spiralling project costs and squeezed profit margins. While the bundled approach of packaging the solutions with tools and services worked for existing clients, many of the newer clients wanted unbundled solutions.
TAIGER also realised that some tools in each bundle could be reused for different application purposes, especially in cases where the client purchased all the product bundles. However, a bundled approach to product delivery and pricing meant that the customer was performing the same task twice and charged twice as well. Such situations forced the firm to reconsider its business model. About 30 percent of TAIGER’s clients were using more than one product and increasingly asked for a varied set of solutions.
Moreover, some clients had their own IT teams who could easily perform customisation of the tools and did not want to pay for the service. They were more contented with paying just for the solution, based on usage. Arroyo started thinking hard on how he could integrate products and yet offer them separately based on the customer’s needs.
Unbundling costs…and benefits?
While a SaaS offering with software and services bundled together had worked reasonably well in the past, the changing market and increased client expectations rendered TAIGER’s existing bundled solutions less profitable. The downside of the bundled model was that it used many resources and limited the delivery of the products to a per-project basis.
Administering customised solutions was also time-consuming and expensive for client organisations, and they lacked flexibility and quick scalability for large-scale implementations. Moreover, bundled pricing meant that clients needed to have bigger budgets for implementation; smaller companies with tighter budgets naturally fell off the radar.
TAIGER had found that demand for NLP solutions was increasing rapidly, and there was a large market for its products amongst smaller organisations. Delivering its products to a larger market segment comprising large and small organisations required a ‘thinking outside the box’ approach.
Arroyo wondered if designing a new business model was the right way forward. Would he also need to devise a new pricing strategy? How could he unbundle the business value of TAIGER’s NLP solutions? More importantly, how could TAIGER deliver on its mission of offering customised AI solutions to a varied set of customers with standardised software?
This is an adapted version of the SMU Case, "Taiger AI: Unbundling the Business Value of NLP”. To see the full case, please click on the following link: https://cmp.smu.edu.sg/case/4846”.
Follow us on Twitter (@sgsmuperspectiv) or like us on Facebook (https://www.facebook.com/PerspectivesAtSMU)
Last updated on 29 Apr 2021 .
