Analytics for Small Business: 4 myths and 2 examples which bust them

Posted by Bala Deshpande on Wed, Aug 01, 2012 @ 09:20 AM

As consultants and analytics practitioners, we believe that there are many ways in whichwizard to dispel small business analytics myths resized 600 business analytics can benefit small business' productivity. But small business owners sometimes think that they do not have the need for advanced analytics because of a few common reasons or what we like to call myths:

  1. They believe that they do not have sufficient data to run any analytics
  2. They believe that analytics cannot really reveal any thing they didn't already know
  3. They believe that analytics will not be able to predict certain scenarios which would have enabled them to be better prepared in the first place

All three reasons may have had some validity in the past, but increasingly these are beginning to sound like excuses for shying away from facing the complex realities of today's business and ignoring the value that other similar businesses are seeing. But there is also a fourth reason or myth, which fortunately is easy to dispel

4. They believe that implementing analytics in a small business like theirs, is not affordable

Here are two examples of small businesses which have dispelled these myths and are succesfully reaping the value from analytics.

Problem 1: Do we have enough capacity for meeting the demand for a specific product?

One of our customers is a company that manufactures anti-corrosion products and sells them to the oil and natural gas industry. They make several different types of products whose demands fluctuate and a challenge is to schedule their production in such a way that they make the best or the most optimal use of their limited production capacity. For example, if the demand for brown tape (one of their products) is low today, but likely to spike in the next few weeks, they want to keep the machine which makes that product ready to go at that time, but not idle today. That same machines can be used today to produce white tape, for example. The key is to determine when to switch production to brown tape from white.

Having a good idea of the demand variation and information about production rates and man power utilization is key to optimize this process. Sales data and production data (number of units, time to produce, cost of materials) must be combined. Predictive analytics enables them to forecast demand using sales data, while a basic analytics dashboard will enable the plant manager to monitor the current production and adjust the schedules to meet the demand.

Problem 2: Manufacturing overhead is a significant cost. Controlling it requires real time monitoring

One of our other customers designs, engineers and manufactures custom equipment for the military. A major challenge for them is to control their manufacturing overhead costs which in some periods can be as much as 90% of their total cost. They need to understand what factors influence overhead, and need to identify events which can add to overhead: such as the hiring of a new IT personnel or changing overtime policies, for example. Clearly a real time overhead monitoring tool would be of great value. However before building such a dashboard, it would be important to define and establish correct allocation methods, by answering questions such as "does assembly time add to overhead?"

The solution for this situation is basically to build an overhead allocation model, dynamically pull data from their accounting system (such as QuickBooks) and post the information to a dashboard, similar to the one shown below.

 tracking manufacturing overhead using dashboards resized 600what if analyses using manufacturing overhead tracker app resized 600

In response to Myth 1, the second example shows that, any business that collects financial data using ubiquitous tools such as QuickBooks has enough information to run analytics. Myths 2 and 3 in our opinion are basically smoke screens to hide behind Myth 4: affordability.

Advanced analytics software has become very affordable today, with many open source tools such as RapidMiner or R. Most of our small business customers are in the $5-$20M range and employ between 50 to 200 people. Investing in advanced predictive analytics should not cost more than a few hundred dollars a month for the first year. Once the basic structure is in place, this cost will drop down even more.

Are you a small business facing any of the above issues? Are you looking to try out affordable analytics? Check out our Analytics Pilot App Program today.

Tags: manufacturing analytics, SME analytics, business analytics