It’s a been a high learning curve year as this has been my first full year working for Acumatica. I have taken in a lot of information, experienced a lot of “firsts”, but I haven’t taken much time to push some of that information back out in the form of blog posts.
This would have been a great year to blog because I believe that the best blog posts are written right after you first learn something, when you still freshly remember the mental barriers that you had to overcome, and you know how to walk in the shoes of someone learning the thing for the first time.
I just didn’t have the energy to post more frequently.
It’s also been a rough personal year. My dad was diagnosed with dementia in December 2015 at the fairly young age of 59. Coming to terms with that hasn’t been easy. I’ve lost a good amount of sleep and added a good amount of weight in the process.
This is a personal blog. I don’t get paid for it and it isn’t part of my duties as an Acumatica employee (note: I wouldn’t have it any other way; I like to keep this thing as a “hobby” and not as a “job”).
Bottom line, the blog has suffered in 2016. Even though I have taken in a lot of information, there haven’t been many nights where I had the energy to get on the laptop and begin the lengthy journey towards completing a post.
But, I’m proud that I was able to post at least once each month in 2016. Except for December. So that’s why I’m sitting here in the dark on the floor in a bedroom, keeping an eye on my 3 year-old son as he sleeps peacefully, blissfully unaware of ERP, Business Intelligence, home mortgages, auto payments, terminal illnesses, or any other annoying thing that us “grown ups” have to deal with.
Here are 8 of my fairly random takeaways from 2016.
1. Data Analysis First, Applications Second
The more I learn about Data Analysis, the more I can’t help but feel that Data Analysis takes precedence over Applications.
Applications are all about helping you to capture data. Or you could say, “data in”.
Data Analysis is all about helping you to analyze the data that you’ve captured. Or you could say, “data out”.
You don’t have to put all of the data into one application in order to analyze it. Especially nowadays with the prevailing “best-of-breed” philosophy, data is always going to be spread across multiple applications. Of course, it’s easier if you only have to go to one place to get the data, and I’m pretty sure that the pendulum will swing back towards “ERP” instead of “best-of-breed” in the next 5-10 years.
But it doesn’t really matter either way, because the Data Analysis process can aggregate data from any number of places.
I always chuckle when I hear people talking about trying to capture data, even though they don’t know what they are going to do with it. It seems like Accountants are especially guilty of this where they try to anticipate every question that someone might have in the future and capture the data needed to answer those imaginary questions. In my opinion, this is backwards. You should decide what questions you want to answer and how to approach the data analysis before even bothering to capture the data in the first place.
So I think that the whole point of capturing data in the first place is to analyze it.
In other words: Data Analysis first, Applications second.
2. The Power BI Gateway Drug
I had no idea what I was getting into when I picked up Power BI a few years ago at the recommendation of my neighbor (click here).
Before I got into Power BI, I only knew about transactional systems like ERP systems which are usually highly normalized relational databases that are designed for getting data in, also known as OLTP (Online Transaction Processing).
Power BI opened up a whole new world for me. First the world of Business Intelligence and Data Warehouses. This is also known as OLAP (Online Analytical Processing). The whole design is for getting data out.
But this Power BI “Gateway Drug” has opened up more than just Business Intelligence on structured data. There are unstructured “Big Data” streams from Social Media, connected devices (the “Internet of Things”), written documents, emails, etc. That’s a whole new territory with a whole new set of tools that I’m starting to explore. I can actually apply the statistical stuff that I learned in college in the real world!
And Cloud applications with their APIs make this so much easier. Cloud APIs aren’t just for transactional systems like Acumatica ERP. They can also be used during the process of Data Analysis. You can use Power BI to perform some ETL work, pass the results to Azure Machine Learning, trigger some kind of action using Microsoft Flow, then pipe the final results back into an ERP product for storage or out to something like Tableau for visualization.
So even the Data Analysis process is a best-of-breed approach, with multiple applications involved along the way.
3. Applications Don’t Implement Themselves
I already knew this one, but it got reinforced this year.
You can buy a really cool new cloud application, but it isn’t going to implement itself.
Especially in the world of ERP where an implementation crosses departmental and political boundaries, the “soft” skills are critical. Business-savvy professionals have the edge over the “geeks” here in my opinion. Managing a project is not easy, but good project management skills are crucial for implementation success.
This is why I really like the 100% channel approach taken by Acumatica. I’m sure it can be a pain at times having to deal with so many partners, but those “boots on the ground” implementation professionals are crucial to the success of each and every application implementation.
Also, I’ve never seen a self-implemented ERP project go well. I’m sure they are out there, but it seems to me like implementation projects that are undertaken without the involvement of a 3rd party consultant are doomed to fail for the most part.
4. Knowledge Getting Cheaper, but Experience Still Valuable
There is so much free/cheap content out there. It is growing by the day. And Google makes it easy to find.
But converting that raw Knowledge into Experience is still an expensive process.
This is one of the reasons why I like to blog. I might be giving away knowledge, but I’m not giving away experience. Besides, the more knowledge you “push” out of your brain, the more space you make for new knowledge.
I refuse to believe that you can break everything down into a formula. There are certain “finer” things that can’t be understood, but can be felt. Like a professional athlete’s ability to get into “the zone”, a place that involves the entire human being, not just the head/brain.
A seasoned ERP implementation professional brings more to the table than just an encyclopedia of knowledge. They have the ability to help guide an implementation into “the zone” because their presence brings a certain level of calmness, they can “feel” problems coming before the data is even there to detect it, and even their fingertips have certain amount of “muscle memory” honed by years of experience.
Which leads me to my next random point.
5. Humans > Machines
Humans are greater than Machines.
I will always firmly plant my flag on the side of humans not on the side of machines.
I like using machines, but I don’t like when machines try to use me.
I emphatically believe in being “data informed”, not “data driven”. “Data driven” organizations will ultimately, in my opinion, become too obsessed with the numbers, alienate their employees in a too cold and calculating company culture, allow metrics to conquer innovation, and ultimately “drive” right off a cliff.
“Data informed” organizations keep human beings in the drivers’ seat, but make more information available at their fingertips. Machines are better at the mundane repetitive tasks, but humans are better at making the important decisions. At least I think so.
Also slipping in here my dislike for self-driving cars. To me, driving a car is one of the symbols of American freedom: the open road, the family vacation road trip, the option to choose the “inefficient” byways over the most “efficient” freeways, the pleasant challenge of navigating the turns of a mountain road. Are we allowed any mundane pleasures anymore or are we being relegated to amorphous blobs where the only muscle that gets exercised is the brain while the rest of us is destined to atrophy.
Machines should serve humans, not the other way around.
6. Dynamics 365
It’s going to take 1-2 years to get a solid idea of where this thing is headed, but Microsoft’s launch of Dynamics 365 has so far been a somewhat quiet affair, yet it will have a big impact on the ERP world for years to come. Lots of questions remain, but it does seem clear that Microsoft is out to commoditize ERP and make it something that you can buy in the checkout line at your local grocery store. It’s an ambitious undertaking and we’ll all be keeping a close eye on it.
One thing I’ve learned though in the past couple of months is that much of the sizzle from an ERP perspective is actually in PowerApps and Flow. I’m not sure yet about PowerApps, but it does seem like Flow can work with any API, including the Acumatica APIs. And it seems like PowerApps and Flow were actually built by the Power BI team, not the Dynamics 365 team, making Dynamics 365 more of a collection of parts, not a cohesive solution. But I could be wrong. Still lots to learn.
7. Serverless Computing
Cloud initially just meant “just another computer”. First you had virtualization technology making it possible to run multiple “computers” on one set of hardware. Then you had big cloud data centers using that virtualization technology. But, for the most part, it seems like much of the “cloud” is just a bunch of bloated virtualized windows servers that have a huge overhead cost for running their applications. The new serverless architecture approach taken by companies like Amazon seems very interesting to me, especially when it comes to Data Analysis where I really just want to work with services, not servers. My hope is that Serverless Computing will be much more efficient and simpler than current operating system virtualization technology.
Speaking of Amazon, I thought it was very interesting to see this announcement from Acumatica (click here). As far as I know, this is the first offering that actually limits the number of users. And you can buy it through Amazon rather than through a traditional VAR partner. Definitely interesting and I would suspect that we’ll hear more about it at the Acumatica Summit next month.
8. I’m Very Thankful
I’m very thankful to be where I’m at in my ERP and Business Intelligence journey. Working for Acumatica has been great this past year. I have learned a lot and I’m sure that I will learn even more in 2017. I just hope that I’m able to “push” more of what I learn out here on my blog.
Happy New Year!