I’m prepared to argue that once a company gets above 2-3M in revenue, or has multiple sites to manage, they need a data warehouse. Why, might you ask? I’ve got multiple reasons, all based on my experience working with or for companies from 3M to 3.7B in revenue, across industries from manufacturing, to healthcare, to technology, to education, to services.

1. Companies these days take advantage of cloud platforms, often with some bespoke/custom platforms, and maybe some aging on-premise or hosted platforms. The data needed to show all the steps from Lead to Order to Delivery to Cash is therefore guaranteed to be spread across multiple platforms, and therefore not easily reportable in comprehensive reports or dashboards. Data warehouses are a destination for critical data from these platforms, with agreed upon business definitions of what each data element means, and inherent time-series storage of data so you can do period over period reporting to see trends. Having this single source of critical data is a requirement these days for getting a handle on all your Key Performance Indicators (KPIs) to effectively run your organization and understand the impact of decisions in marketing, sales, contact centers, operations, HR, distribution, inventory, finance/accounting, and collections. Trying to bring all this data together using spreadsheets is the normal practice for most small (and, sadly, some mid-size) companies, and that just is not scalable while also being prone to human error.
2. Data warehouses time series data storage is key for detecting trends because often sales, marketing, operational and financial platforms do not store data about when various events happen along your Lead-to-Order-to-Cash process. Those platforms instead store the current status of events, not the full history of events. Even if they do store the full history, often it isn’t easy to access, nor in a format that supports reporting/graphing. Lastly, sometimes the security isn’t adequate and users can change, mess up, or even delete the history without realizing the impact. Every business needs to see and understand trends so they can react to them in a timely manner. For example, if sales, orders and cash receipts are strong right now, but you don’t know that your lead volume has dropped precipitously because your order volume happens to be high, you will very easily miss the fact that you need to fix or juice something in lead generation until you notice that orders are dropping. That’s too late to react, and you will have a downturn in orders for quite some time until you fix the lead process, versus reacting sooner and either fixing the lead process, or at a minimum minimizing the downturn timeframe before recovery. Connecting your platforms to a time series data warehouse, and feeding it every time critical data changes during it’s life cycle, is the key to having data available to show you exactly what is happening at all steps in your business cycle, thereby allowing you to analyze those steps and take action more quickly.
3. As companies get bigger they have more executives that want to track data using their own definition of terms, timeframes, status, and much more. This inevitably leads to disagreement on who’s numbers are correct, when the real answer is they all are representations of different definitions of what ‘correct’ means. That is a recipe for executive infighting, poor decision making, slow reaction time to change, and, worse, improper reporting of business performance to the board, investors, etc. Data warehouses force definitions of data that everyone has to agree to, with clear tracing of where the data came from (e.g. which platform) and how it was changed or transformed or calculated to end up in the data warehouse. Even better, if executives decide to change a data definition of something in the future, say, like the definition of what qualifies as an approved order, you can change that definition once in the data warehouse and not worry that all kinds of other data sources or spreadsheets haven’t been updated with the new definition. Even better, you can put everyone’s different definitions into the data warehouse, but with clear labels for what they each mean, how they are calculated, and what they represent when viewed by a user. This clarity ensures people report on the right things based on the right consistent set of definitions and understanding of what the data represents, and therefore tells them about the business.
4. The best reporting and dashboard tools today, such as Tableau and PowerBI, work best when they have a data warehouse with time series data. These tools don’t work well when you have to connect them to 5+ platforms with different states of data with different definitions and missing historical data on changes over time. Therefore, to get the most out of your Tableau investment, having a data warehouse will enable you to build both simple and complicated repots and dashboards that encompass the major parts of your business and allow you to view the business as a whole, not as a series of departments, divisions or products.

There are more reasons than the 4 I’ve listed above to build your data warehouse, so I welcome feedback on any critical reason’s I’ve missed. I’ve yet to find a good alternative that addresses all the concerns I mentioned above, other than trying to standardize on a single ERP that has modules for every piece of your business. However, I’ve never seen a single ERP work for any business, because it’s usually too expensive to buy all the modules (HRIS, SFA, CRM, GL/AP/AR, Operations, Call Center, Field Service, Reporting, Dashboards, Time Tracking, etc). Not to mention, usually your chosen ERP has modules that are simple/basic, and certainly aren’t best-in-class in capabilities. No surprise, users always want best-in-class or best-in-industry platforms for various modules that run their area of the business, so they inevitably pick different software packages that then have to be integrated with the ERP, and usually only integrated minimally to save costs. 

If you are interested in who can help you build a data warehouse, I’d recommend a company called Cleartelligence (www.cleartelligence.com) who specializes in this kind of work. I’ve worked with them at multiple clients across healthcare, services, and other industries, and they are the real deal. For full disclosure, I do work with Align Capital Partners who owns them, and that’s how I was introduced to them. However, for me, they have a proven track record of 100% success where I have recommended them, so they are my go-to company to help organizations build their first data warehouse or enhance/expand their data warehouse.

Thanks, Eric Dirst, President/Owner DeKonsultere LLC