With Textmetrics, organizations can centrally manage their online and offline text outputs and continuously improve their effectiveness and consistency. We use machine learning and artificial intelligence (AI) to provide real-time text suggestions that appeal to the target audience. For example, significantly improving the conversion rate of job advertisements.

Obstacles

When developing a SaaS service, several challenges arise. The biggest is not specifically a challenge for SaaS, but it plays a role here. It is the challenge that you must develop according to customer needs. This means that at the moment you have an idea, you do not yet know if it is a good idea. Many companies struggle with this. You must develop what the customer needs, while you tend to think that what you are developing is good. Sometimes this also means that your idea is not as good as you thought. And then it is simply a matter of accepting it and starting over.

The needs of an organization like NCOI differ from a customer like ING. You must test every functionality. After which you ask the customer: will you use this? Will you pay for this? Only when the customer answers a resounding yes, do you start working on further development. But you do not develop the product too far, because you first need more customers who also want to buy the product. Then you are talking about a minimum viable product: a product with just enough features to add value for early adopters, with the goal of getting feedback for further product development. A product is only viable if customers are willing to pay for it. If that is the case, then you will further develop the functionality.

Another major challenge in developing a SaaS service is that the costs come before the benefits. Even when developing other products, you as a company must make an investment for the customer in advance, but then the customer will pay for the product you develop. With a SaaS solution, you build a product for tens of thousands of customers, while your first customer does not pay nearly the total amount. You must invest a lot in advance and only get paid later. You need to keep this in mind when building a SaaS solution, as you need to attract enough paying customers later.

Customer Orientation

At Textmetrics, we tackle the above challenges by developing with a customer-oriented approach. We work according to a lean startup methodology. We ask beta users to test our products and develop our products based on specific questions from the customer. We always ask ourselves the following questions: is the functionality the customer is asking for broadly applicable to other customers? Or is it a specific functionality that is only interesting for this one customer? In the latter case, we will do it, but it is truly custom work. We develop our SaaS service based on what the customer wants, not based on what we think is good.

Focus

Technology is becoming increasingly complex. The number of cloud platforms is increasing, and artificial intelligence (AI), big data, and machine learning offer more and more possibilities. Everything is becoming extremely specific. Therefore, you really need to focus on what you are good at. In the future, it will become increasingly difficult to develop a broad SaaS solution that can do everything. Therefore, it is better to choose a more specific solution and collaborate with other parties for broader solutions. Focus entirely on what you are good at.

Carefulness

One of the questions raised about texts that are created with the help of AI is whether those texts truly meet the needs of the customer. And whether they primarily serve the needs of the company. Because is the customer being served with what they want? Or does the customer choose based on how we manipulate the text?

In short: AI should not pose a threat to freedom of choice. Therefore, you must handle AI learning carefully. This prevents texts from being written according to a certain structure and with only one goal in mind: to convince the customer. Aside from the fact that it is currently not possible to create a complete text using AI, you should not solely rely on machines. It remains partly a human endeavor.