R&D in a startup: how to create innovations for the needs of the audience
While Research and Development (R&D) does involve innovation, it does not always involve IT and engineering technology. In fact, this direction is focused on market research and creating products for its needs. In the fashion industry, R&D teams are working on new images, and in the restaurant business, on new dishes: for example, KFC has its own Research and Development department.
The opinion that only large corporations can afford R&D is nothing more than a stereotype. This direction is available to businesses of any size - the only question is the size of the team and costs. Tech giants are making huge investments in R&D: Facebook has 21% of annual revenue, Alphabet has 15%. In smaller companies, this figure can be much lower - unless the company's specialization dictates otherwise. The volume of the budget that goes to Research and Development also depends on many other factors: the company's economy, business margins, the role of innovation in it and the industry. So, if we consider the ratio of R&D expenses to EBITDA (profit before interest, taxes and depreciation), in the consumer sector this figure reaches 32%, in banking - 12%, in the IT industry - 42%, and the highest in pharmaceuticals - 52%.
Any R&D task begins with studying market needs and generating ideas, and ends with the development of the product itself. Therefore, such a team should include both developers and engineers responsible for creating a solution, and analysts - those who determine how cost-effective it will be and formulate the business requirements.
Typically, the product team consists of analysts, product managers, engineers and developers, a development manager and a structure manager. But to “take” them exclusively for R&D initiatives or not depends on the specifics of your business processes. Many companies practice a hybrid approach: when one employee can be engaged in both the design of new products and other tasks. And in large corporations, R&D is a large division that focuses only on market analysis and development of new solutions.
Another approach to an R&D team is to outsource research issues. In Russia, this method is often used by industrial corporations that entrust the creation of new solutions to scientific centers such as Moscow State University, MIPT, MEPhI, Skoltech. But in this case, it is difficult to manage the development process entirely and completely, so many companies in parallel develop their own expertise. This is what the Sibur holding, for example, does: the company gives the initial stages of R&D to partners, and conducts the final stages independently.
If you are making R&D solutions not for internal use, but for customers to order, another approach to team building is possible - to combine internal specialists with employees from the customer's side.
Generating hypotheses and testing them
The R&D cycle starts with generating ideas. To do this, product managers research the market, study existing products in the world, analyze their pros and cons. If there is an obvious problem, but there are no solutions for it or they do not suit you (due to cost, difficulty in maintenance, and other factors), the concept of your own version begins. The development team is engaged in this: the design department, IT specialists, in the restaurant business - the chef.
Sometimes R&D activities are based on existing solutions - when you need to adapt a product for another market or create a new one for it. For example, the Russian R&D division of STADA PharmDevelopment is engaged not only in the release of new drugs, but also in support of existing ones to ensure their compliance with the changing requirements of pharmaceutical regulation.
When an idea for a new product is formed, it needs to be redirected to business analysts: they will assess how much it will cost to implement, how quickly it will pay off, and whether it will pay off at all. At this stage, it may turn out that the development of the product is not profitable - then you will either have to change the concept or abandon it altogether.
An obligatory component of the development stage is an audience survey, or customer development. A product is never created on the basis of companies' assumptions about its relevance - it must meet the real needs of the audience. Therefore, before starting development, it is important to communicate with your customers to find out if they are ready to use this solution.
One of the main methods of initial customer development is in-depth interviews. Later, when you launch the MVP, it will be possible to conduct tests - but until you have only an idea, you will have to talk a lot with the audience. And at this moment it is important not to make a common mistake - in the process of communication, do not try to "lead" the respondent to the right thought. Wordings like "Do you like ...?", "Would you use.?" unwittingly push the interlocutor to agreement, so he is likely to answer in the affirmative - but the conclusions are distorted. So tryavoid closed questions: instead, it is better to be interested in what a person wants, what his needs and problems are. Try to get as much information as possible from the interlocutor: it will help you avoid many mistakes and save money, time and nerves.
From prototype to pilot
When the concept has passed the test, it goes back to the development team - this time to create an MVP. It is tested on an audience, then adjustments are made to the product based on the test results - and testing is started again. There can be several such cycles - and only after the final revision does the pilot start.
Often at this stage, many are carried away by endless "improvement", striving to first make an ideal product, and only then show it to people. This is a serious and, unfortunately, a common mistake: this way you run the risk of doing a lot of wasted work - if during testing it turns out that the solution has many flaws or is not needed at all.
To avoid this scenario, it is important to test the hypothesis with a real audience as early as possible - and repeat the test after each iteration. Do not be afraid to show the raw product: if the idea is really good, it will "go" even with a bad design and inconvenient interface. Your task is to identify all the flaws and fix them with minimal losses. Therefore, the more often you conduct testing, the less time and money will be spent on development.
Estimates and deadlines
After the introduction of a new product, it is time to evaluate how successful it turned out. This can be done only after its implementation, and the key assessment metric will always be the payback of the solutions created. Any innovation should help you either earn more or spend less. For example, in 2018, Cognitive Pilot (a joint venture between Sber and Cognitive Technologies) developed an autonomous control system for a combine harvester. For the first time, this system was used by enterprises only two years later, in 2020 - but it immediately showed its effectiveness: unmanned technologies helped to save more than RUB500 million on grain harvesting.
If the solution does not bring profit, it is a signal that something is wrong with it. This means that you need to restart the R&D cycle: conduct audience surveys, form hypotheses, make changes to the product and test them - until the output turns out to be something that will close customer requests and bring the company a profit.
As for the timing, the entire R&D cycle, from the emergence of a hypothesis to the launch of a pilot, can take from a couple of months to several years - it all depends on the complexity of the solution and the industry. For example, in pharmaceuticals it takes about ten years on average to release a new drug. In engineering and IT companies, this period is usually much shorter - if, of course, we are not talking about global tasks, but about products for a specific request. Moreover, most of the time is spent on the initial development.