DataHub Consulting, Experts in Analytics, Business Intelligence, and Compliance 1200 627

Written by

Date

4 February 2025

Category

In this article we are going to be discussed Microsoft Fabric and particularly, Why should small and medium sized businesses use Microsoft Fabric. As a data consultancy, Datahub speak with potential customers about what data they have in the business, their current methods of storing, processing, and reporting of data, and the data challenges they face.

As a solutions provider we will suggest solutions that in our opinion will be best suited to the customer’s needs. In doing this we will look at the financial cost of development and year on year running costs, as well as the technical aspect of meeting the customers’ needs and addressing any data challenges.

When we suggest using MS Fabric as a solution, we have heard initial comments of, ”we are only a small business, is Fabric overkill”, or “I’ve heard that Fabric will cost us thousands each month”. I get the impression that there are some misconceptions and myths surrounding MS Fabric. After we discuss the concerns with the customer it appears that they have prejudged the technology, or have been given some incorrect information. Yes, MS Fabric can cost thousands of UK pounds per month running costs but for majority of SMBs then this is not the case.

So, Why should small and medium sized businesses use Microsoft Fabric. Let me explain…

What is MS Fabric

MS Fabric is a Software as a Service (SaaS). It encapsulates data engineering, data warehousing, Real-time analytics, and data science into one platform. In addition it also includes Microsoft services and technology like Data Factory and Power BI. Making Fabric an end to end data solution for businesses of all sizes.

There’s one platform, one resource, one subscription, and one data lake to support data engineers, data scientists, and data analysts. OneLake with delta tables is at the centre of the Microsoft Fabric ecosystem.

Allowing all technical teams developing from the same platform allows for a faster development cycle and reducing the cost of delivery and increasing the ROI. For businesses with multiple developers Fabric also allows for CI/CD with the integration of Azure DevOps and Github.

Having spoken to an SMB that migrated from Azure to MS Fabric, the business has made a 45% saving on monthly costs.

What is an Small Medium-size Business?

Image showing a development meeting in progress

SMB are Small and Medium-sized Business, and this type of organization make up over 99% of the businesses globally. In parts of the world SMBs are also referred to as Small Medium sized Enterprise (SME). In the U.S. they use SMB more frequently, but in Europe they use the term SME.

These businesses are a key part of the global economy, creating jobs and driving economic development and innovation. For this reason, SMBs need to be data driven and have control over their data.

Global Number of Small and Medium Sized Businesses

United Kingdom

Data from the UK government statistics on business population estimates that there are 5.5million SMEs in the UK in 2024, equating to 99.1% of the business population.

https://researchbriefings.files.parliament.uk/documents/SN06152/SN06152.pdf

European Union

Reported by Statista, the European Union (EU) has an estimated that 26.1 million small and medium-sized enterprises operating in the EU borders in 2024. This includes all 29 countries that make up the European union.

SMEs in the EU 2024, by size | Statista

United States

In 2024 there are an estimated 34.8 million small and medium sized businesses. This is an increase of 4.5% on the previous year. Equating to 99.9% of businesses currently operating in the U.S.

New Advocacy Report Shows Small Business Total Reaches 34.8 million, Accounting for 2.6 million Net New Jobs in Latest Year of Data – Office of Advocacy

Case Study Example

This is a case study of a small business that Datahub Consulting have supported with the implementation of MS Fabric. We will refer to this case study through this article as an example.

A small manufacturing business with approx. 40 employees that manufacture plastic parts for the automotive industry. They don’t have a data development team to create a solution but employ an analyst that is good at creating and delivering reports with charts and visuals. This analyst has previously used excel to report to the management team and doing any bespoke analysis as required. This business has acquired another business and so there are two different ERP systems in place, one for the business and one for the acquisition. Bringing together data from both ERP systems is required for reporting. This is primarily to monitor and manage stock control of part, supplier information, customer distribution, and orders per month.

The business decides to get Datahub Consulting in to review the data sources and to recommend a solution to allow for reporting across the whole business including the acquired business data.

To keep the solution simple and cost effective the recommendation was a solution using MS Fabric and Power BI. Based on usage and the volume of data it was recommended to have an F4 capacity with 6 Power BI licences for users that either develop reports or users that consume and view the reports. The management information was collated by one person who created and updated a management pack once a month for a management meeting. There was also PowerPoint slides that utilized visuals from reports and Datahub shown the team how to effectively ingest visuals into a PowerPoint template. Any additional or bespoke reporting requests were handled by the analyst who created and presented back any reports or numbers.

Cost to the Business
MS Fabric F4 capacity would be $613.20 as a pay as you go (per month), but with a yearly commitment the reservation cost would reduce to $364.67 (per month). This is based on the region being UK South. For reference at the time of this article comparing regions, If the capacity was hosted in Central US instead of the UK, then this would equate to a 14% reduction per month. There are benefits to hosting the data in the same region as the business but it’s not always essential. We recommended South UK as it would impact latency if the data was hosted in a U.S. data centre. Also, if the data contains personal information and data protection laws apply there would be a business case to host the data in the same region as the business.   

The Power BI PPU (Premium Per User) licence for 6 users would normally be $20.00 per user per month. As the users already have Office 365 E5 licences then this is reduced to $10.00.

The total cost of licencing for Fabric and Power BI would be $5,096 US Dollars. As it’s a UK based business this would equate to £4,138 UK Pounds.

Please note that all licence costs, and US dollar to UK pound conversion rates are accurate at the time of this article. As of April 2025 the Power BI Pro licence will increase to $14 per month and the PPU licence will increase to $24 per month.  

Why use Microsoft Fabric?

The business had a growth plan over the next 5 years that would see them move their offices and manufacturing to a new building. Increasing the manufacturing capability and allow for increase in the number of orders per year. This would also mean increasing the employed staff. With their current state and the growth plan, creating a fit for purpose business intelligence solution that will scale with the business. Using the F4 capacity could then be increased as growth and data increased.

Myths

Having spoken with some CTO’s and CIO’s there are some myths and misconceptions of MS Fabric. Let us have a look at some of these that we have come across and explain.

  • Microsoft Fabric is to Expensive
    Some people think that the F64 is the base price and instantly say that’s to expensive. This is because the F64 is the equivalent to the Power BI P1 capacity and has been publicised for comparison figures. For some SMBs the cost of an F64 is too expensive, I understand that. But they don’t realise that MS Fabric starts at F2 which could be as low as $182.33 per month which is more than affordable to 99.9% of small and medium sized businesses. Having a low capacity means that there is a low data threshold and if the capacity is not sized correctly could incur throttling. But with the correct sizing any overuse will be consumed with the bursting and smoothing feature that we will explain more about later in this article.
  • Fabric is only for large enterprise organizations with 100s of staff.
    The answer to this is definitely not.
    As I started to explain in the case study there are use cases where small and medium sized business are using MS Fabric at an affordable price. Yes, it can scale to meet the needs of large corporations that will have many capacities. But the starting point for capacities does cater for the small business in mind. As mentioned in the previous paragraph it will require correct sizing for the business including the volume of data, the amount of processing referred to as CUs consumed, and the number of users that you have creating and consuming reports.
  • Is Fabric only a database.
    MS Fabric does include serverless databases but is more than that. It’s a complete data platform providing scalable applications with security and data governance. As a managed service Microsoft look after the resources that previously the business managed in Azure. The development team would create a workload like a data lake, data warehouse, database, real-time intelligence etc, and Microsoft would provision it in minutes. Yes, settings would need to be reviewed but Microsoft do all the hard work provisioning the workload. This means the number of technical people required within a business to maintain the platform is reduced.

What is a CU and why is it important.

In MS Fabric the storage and compute are split out. With any cloud provider storage is very cheap. The term CU refers to “Capacity Unit”. A capacity unit is a measurement of the compute power available for the particular capacity. The higher the capacity the more CUs available. So, in simple terms each time you ingest data, transform and cleanse data, run a report that queries the data, all of these actions will use compute power and will consume CUs.

Microsoft Fabric Pricing Structure

The pricing of Fabric is based on a capacity size. All of the capacities are prefixed with an “F” and start at F2. There are a range of capacities available to purchase. With each capacity Microsoft invoice on a monthly basis regardless of the size of the capacity. With a yearly commitment the monthly cost will reduce by approx. 41%. This means that you still pay for the capacity on a monthly basis, but you give Microsoft a yearly commitment to purchase. Some businesses use the pay as you go price to evaluate the capacity needed and then make a yearly commitment.

At an organizational subscription level, previously with the Power BI capacity it was made up of “P” (premium capacity) SKUs, and the cheapest option was a P1. There was an individual licence called “PPU” (Premium Per User) that was a cheaper option, but to compare like for like with MS Fabric need to look at organizational subscription licences only. For comparison a P1 capacity is the equivalent to an F64 Fabric capacity based on v-cores.

Here’s a link to the Microsoft page where you can find up to date pricing for Fabric capacities. Select the region where you plan to have the capacity and the currency of your choice.
Microsoft Fabric – Pricing | Microsoft Azure

Trial Capacity
For reference there is a trial capacity for Microsoft Fabric that will last 60 day. There is an option to extend the trial but this is only designed to evaluate the functionality of Fabric. The trail is the same size capacity as the F64, so if you then purchase a lower capacity such as an F16 then the performance will be lower due to the difference in v-cores. I would not recommend implementing the trial into production as the trial facility could be stopped at any time by Microsoft. I would only use the trial for POC (Proof of Concept) purposes.

Benefits to the Small and Medium Sized Business

For an SMB there are many benefits to using MS Fabric. By now, reading this article you will have some idea of the benefits but let look at these in more detail.

Scalability

MS Fabric can scale depending on the size of the business or the requirements of the business. Yes, MS Fabric can support large enterprise organizations and has additional features to support this once that you get to F64 or above. For example:

  • Businesses can use larger semantic models.
  • Users consuming reports can do this with a free viewer power bi licence.
  • Copilot can be enabled.

But for the smallest of businesses fabric is just as beneficial. The vast majority of features available to large businesses using a larger capacity are also available to the smaller business using an F2 or F4 capacity.

Also as described in the use case, as the business scales and increases then the capacity can do so as well.

Even for start-up businesses having a small capacity (F2 or F4) coupled with Power BI either Pro or PPU licences can provide an effective data management and reporting solution.

Using the OneLake to store data whether it be csv’s, excels, structured data in tables, or files containing images etc. In OneLake it makes use of Delta tables that are an effective format for reporting.

Direct Lake Mode

Direct lake is a storage mode used by semantic models. Previously Power BI models used Import, Direct Query, and composite storage modes. With MS Fabric this allows for another storage mode called Direct Lake. Fabric stores tables in a delta format which are parquet files with delta logs attached. This is a very efficient way to store data.
Delta tables are known for their high performance due to optimizations like data caching, indexing, Z-order clustering, and efficient data layout, which significantly improve query speed and reduce latency. With Direct Lake mode semantic models connect directly to the delta tables giving performance improvement over Direct Lake, and improve latency over Import mode.

Integration

Within a development team there can be one or more technical people. This could include data engineers, data analysts, data scientists. Usually, these technical people work in their own environment creating their own development deliverables. This can mean duplication of data, time and effort, and resources. Microsoft Fabric offers significant integration benefits by seamlessly combining various data services and tools within a single platform, allowing for unified management, governance, and data discovery.

From a coding perspective, The MS Fabric platform provides a solution that can adapt to the skills and knowledge of the technical team. Firstly, for developers with no coding experience there are low code / on code methods of working.

For data engineers that are experienced in development code Fabric also allows for integration of different programming languages supporting SQL, PySpark, Python, Scala, R. Integrating both the Spark and SQL engines for performance.

Collaboration

MS Fabric allows the whole business regardless of department, region, or job role to collaboratively work with the same data using the same tools, and same reporting solution (power bi). You may have come across the term “one version of the truth” and this is what fabric gives you.

There are too many occasions where business work in silos and have their own data, with their own reports, giving their own numbers. There could be a discrepancy between the numbers finance have to what the operations team have. Which is correct?

Using MS Fabric OneLake is a centralized data storage allowing for centralized collaboration of data, resources, and cost.

Bursting and Smoothing, what is it?

There will be times where a business uses more CU than their capacity allows. This does not stop MS Fabric. Microsoft realize that this happens and have a feature that can address this called bursting and smoothing.

When a business uses more compute than the capacity allows then this is referred to as “bursting”. Microsoft then will allow the capacity to even itself out over the next 24 hours. This is called smoothing. So, for example if a business does not plan there processing correctly at month end there could be a lot of data ingested, data transformations triggered, and reports used, this could involve bursting. But compute is not free, and the additional compute used will need to be pulled back over the next 24 hours.

The capacity usage is broken down into to categories, background percentage and interactive percentage. Bursting happens when the compute in the graph goes above the 100% capacity with the combine background and interactive percentages.

Interactive % represent the number of CPU seconds used during interactive operations in a 30 second period. This operation covers when Power BI users interact with the report and queries run to generate the visuals. Also interactive % involved when data warehouse views are executed as part of a reporting solution.

Background % represent the number of CPU seconds used during background operations in a 30 second period. Background operations cover Power BI backend processes that are not directly triggered by users, such as data refreshes.

Please be aware that in the case study example the capacity usage has not been planned well, and the bursting and smoothing is used. This feature should not be used in any capacity planning. This is a feature that is used as a last resort option for when any unexpected compute is required. If a business says we have additional compute that can be used and factor this into regular processing, then there will be occasions where the usage means smoothing can not happen over the next 24 hours and when this happens throttling will occur. When throttling happens users will have restrictive usage including locked out of reports and scheduled tasks like ETL processes will fail to run etc. So be warned!

What are the options if bursting happens regularly.

Increase the capacity.

Regular bursting and smoothing could mean that you have outgrown the capacity and require a bigger one. There is a capacity monitoring tool that can be used to understand usage and what queries are using up the CUs.

Optimize your queries / workloads.

With any data management solution there will always be queries referencing the data. For simplicity let’s say some DAX code for measures required on reports. If you have inefficient queries or workloads within the system this could increase the compute required.

Distribute your workloads depending on usage patterns.

Where possible evenly distributing any workload will optimize the compute and provide a better solution to the business users. So, look at:

  • ETL loading strategy.
  • With semantic models use direct lake and not import mode.
  • Utilize shared semantic models.
  • Look at the frequency of when users use reports and dashboards etc.
  • Limit the use of Analyse in Excel. This is a feature that users can use and I’ve seen businesses where this feature is overused. This was the cause of CU limits being exceeded and bursting happening.

Capacity planning or capacity monitoring are large topics and to big to include in this article. For that reason, planning and monitoring techniques are out of scope of this article. But if you require any capacity advice, then contact Datahub Consulting and we would be happy to help.

In Summary

In answering the question, Why should small and medium sized businesses use Microsoft Fabric. MS Fabric can be scaled and provisioned for any business regardless of size and data demands. Fabric is a cost-effective solution based on licencing, cost of delivery, and Return on Investment (ROI). One client has reported to us that migrating from Azure to Fabric has resulted in a 45% decrease in costs. So, there are financial benefits.

There are even prebuilt industry solutions within Fabric for Retail, Healthcare, and sustainability solutions. I see these working for medium and large sized businesses so these do work for a percentage of SMBs.

As there are so many services, workloads, and development options within Fabric a solution for any business needs to be on a case-by-case basis. With a company like Datahub Consulting architecting and delivering a solution would mean that you have benefits from a cost-effective data platform that can adapt to your business needs and growth.

If you want to understand more about MS Fabric our team at Datahub have data engineers, data scientist, and architects that can support. Collectively we are experts in the MS Fabric platform. Myself I’m a Microsoft Certified Trainer (MCT) that specializes in MS Fabric, Power BI, Semantic model optimization, and capacity optimization.

Working with Datahub Consulting



Datahub are experts in delivering data solutions with global customers. Regardless of if you have data teams within your business, we can add value and success. Either delivering end to end solutions, or supporting your internal team to deliver success.

It wouldn’t cost you anything to start a conversation with our CEO that has personally worked with customers on MS Fabric projects.

Contact us
Contact us | DataHub Consulting
Datahub Consulting Website
Data Consultancy Services | Datahub Consulting

Find out how we can help

We do not employ salespeople; our team are all experienced technical specialists that can talk you through any of our services.

Contact us