It's very common for software applications to be implemented without carrying out a data modelling exercise . Usually a lift and shift installation which can turn out to be inefficient and costing the business more than anticipated. 
It's a case of what works for one may not necessarily work for another. If it works for company A, it could also work for company B as long as the software is tailored to Company B. It is important to tailor (model) software to the business requirements.
So why do businesses overlook data modelling? The answer is likely a lack of awareness where the business believes it's saving cost by descoping a data modelling exercise.
I built my first data model about a decade ago when a client engaged me on a project which was intended to select a Business Intelligence tool suitable for its needs. 
We started a Proof of Concept with the choice of two well known Business Intelligence applications. 
The client requirement included prepping data in a number of ways which the data warehouse wasn't able to satisfy. That is, without extensive extra work being done.
We were short on time and a decision needed to be made within the project timeframe. 
The Datawarehouse team, based offshore and still tasked with the "business as usual" tasks did not have the scope nor capacity to support this requirement.
This is a common occurrence. A client has state of the art data warehousing solutions but the business cannot get the information required in a timely and efficient manner. 
I therefore had to step in to ensure that the business was able to understand the benefit of the tools and make an informed decision on which tool would best cater to the business requirements.
The alternative was to conclude that the tools were unable to deliver the required objectives and accept the status quo. This would have been a disappointing outcome for the client due to the costs incurred to this point.
So I set to work, reaching to my knowledge of database design, data structures, facts and dimensions to build a data model which will help the tools show their capabilities! 
Over multiple sessions, I gathered adequate information from the stakeholders to understand what the business processes were, who did what, where and when. I enquired about the metrics required and the respective definitions across two business domains limiting the scope of the project to the core project and resource domains.
It was important to understand the data from the point of view of the stakeholders. What information did the client want to extract from the source systems, what type of data, how was the data stored? Thankfully a lot of the data was structured and there was capability to connect directly to the source. This helped translate the data and model the data in a meaningful manner for the client.
The client was keen to see the tools deliver various functionalities which would excite its stakeholders and show a positive engagement with the business data further. 
Being data driven was the ultimate objective. 
The visuals were important but what was more "pertinent" was the data behind the visuals. 
Of course, certain time intelligence metrics also needed to be developed into the data model to give a better understanding to the client.
It was an interesting an engaging project which delivered results the client was pleased with and more! The data warehouse team eagerly wanted to know more about how the data was able to be presented in such manner. The key was the data model! 
My conclusion was that the existing data warehouse design had not been modelled to the that particular business or alternatively, it was time for business transformation and a move to a new and compatible data platform for my client. Data modelling is essential to deliver trusted insights into your data.
I believe every business should have their data modelled to ensure the right information is being obtained form it's data asset.
Looking forward to learning more about data modelling and various data modelling tools and techniques at the Data Modelling Zone!