Abstract

There is no question that Business Intelligence can play a significant role in increasing internal productivity as well as competitiveness. To some extent I am going to play the role of devil’s advocate in this article and challenge people to utilize Business Intelligence at the right time and by the right people.

Why do firms suddenly need Business Intelligence?

I have read a number of articles recently that advocate the use of Business Intelligence by many people within an organization. Actually the reasoning seems to be valid. IT departments are swamped with requests for service by more and more people. Many of these requests are Business Intelligence oriented. People see things happening in their market place and need to respond as quickly as possible. Business experts say that we need to become more agile, ready to change our operational and sales strategies in an instant to meet an ever more chaotic competitive environment. Business Intelligence tools are becoming more and more powerful.

The recommended solution is simple. Teach people how to utilize Business Intelligence tools so they don’t have to wait on the IT department several weeks. The underlying assumption is equally simple. If people know how to use Business Intelligence tools, they can react quickly as market conditions shift.

Do you really need to move this quickly?

I don’t think that Business Intelligence for SMBs and SMEs needs to be based on the assumption that we need to react quickly; almost it seems as if there’s a panic. If we don’t address the crisis de jour, our opportunity to meet the challenge will pass in just a few weeks. That’s what some industry experts seem to be saying. Be ready to react instantaneously and in order to do so you need instantaneous information from your Business Intelligence system. Since IT cannot react that quickly, it’s up to the individual to generate the information they need. That means these individuals need to know how to use the Business Intelligence system.

I accept the fact that firms need to react to changing market conditions, but I don’t agree that reaction times are measured in days, not months. Business conditions do change, but you shouldn’t try to react with a speed of a commodities trader. Yes, some firms make money in the commodities market, but there are also far too many examples of firms and unsupervised individuals making bad bets.

Maybe the underlying issue is the whole notion of reacting to changing market conditions. If you need to react quickly, that’s a knee jerk motion and you may not have all of the information you need simply because you tried to use your Business Intelligence system to react quickly.

Maybe you should be more proactive?

Rather than trying to use your Business Intelligence system quickly in order to react quickly, you should proactively use your Business Intelligence system to put you in a position where you can react quickly and effectively. I think that’s the issue. You cannot achieve long term success if you are constantly reacting to changing market conditions.

This doesn’t mean you should never react quickly. Sometimes you need to do so, but the key issue is creating a proactive Business Intelligence system so that when conditions change, you already have simulated the conditions. Proactive means that you have discussed various scenarios and have the information at hand when and if conditions change. Actually you should have already defined market scenarios and used your Business Intelligence system to help you determine what your reaction should be. Proactive means that you have already asked and answered the question.

Who should generate the intelligence?

If you have decided that it’s best to be more proactive, who should be responsible for generating the information you need to make sound business decisions? I am still not convinced that people need to be spending their precious time designing Business Intelligence reports. Many Business Intelligence systems are very powerful, but this power tends to be wrapped in complexity. Most people (and firms for that matter) need to concentrate on what they do well. Sales and marketing executives should spend time determining what the firm should do, not determining how vital information can be extracted from the Business Intelligence system. That should be the responsibility of the Business Intelligence guru. Sales and marketing executives should be using information effectively, not generating it.

Does IT need to be responsible? Possibly or you could create one or more super users that know the Business Intelligence system intimately. Business Intelligence for SMBs and SMEs (given their rather limited number of employees) has to be more of a one man or woman show. That’s not the best scenario due to the risk of losing that valuable person, but then that’s an issue for all SMBs and SMEs and every one of their most talented employees. The key question you need to address is in essence the experience/benefit ratio of any Business Intelligence system. People who use a Business Intelligence tool infrequently are going to take longer to extract the information simply because they need to refresh their memory before they begin. At the same time their reliability of the data extracted might not be as high, again simply because they are not quite as knowledgeable as someone who does this more frequently.

In the end there are two objectives you need to keep in mind. If you proactively take the time now to determine what information you might need in the future and create the decision support tools (reports as well as what if analysis tools) you will need, you won’t have to react in a crisis mode when market conditions change. Second, you need to determine whether it makes sense to train people on a complex Business Intelligence tool when they may not need that tool except on an infrequent basis. Again, if you take this step now, you will be ready when the need for information occurs.

Creating an internal Business Intelligence system

Business Intelligence for SMBs and SMEs isn’t just about reacting to outside market influences. It can and should be just as important to control internal business operations. Actually this is the case for any firm.

While there is no doubt that sound business decisions are based primarily on the person making the decision, these decisions need to be based on sound information. Seat-of-the-pants decisions might work for small firms where the experience of the primary decision maker is the key element, but there is just too much information spread over too many decision makers for this practice to work in larger firms. In this case accurate and timely information must serve as the foundation upon which sound decisions can be made.

Why do you need this information?

Accounting and ERP systems can store and regurgitate just about anything you want. Actually it’s possible that for every transaction posted there can be 20, 30 or 40 fields that will allow you to slice and dice data any number of ways. The fact that you can carry reporting (intelligence?) to this level of detail doesn’t necessarily mean you should do it.

Start by asking yourself why you need the information. If it doesn’t lead to a decision, then you don’t need it.

In what format should this information be presented?

With the exception of audit trail reports, rows and columns are out. In no case can they lead to a decision.

Pie charts and bar chart are out as well. They, like row/column reports, are no more than a snapshot of a specific condition at a specific instant in time. People should never make snap decisions and that’s all this information can support.

Having said this, there is one use to which some specific data can be applied. Exception Management or Business Alerts are usually triggered by a specific value, but only if that condition falls outside an expected range. As an example, if a customer’s account exceeds its credit limit or a specific invoice becomes “x” days overdue, that condition should be brought to the attention of a named individual. Business Alerts handle the notification process while Exception Management gives users the ability to deal with an alert in a contact manager like application that allows them to track the steps they (and others) take to resolve an issue.

If rows and columns, bar charts and pie graphs are not acceptable, what’s left? Line charts are the only form of reporting that actually lets users develop a sense of the history and future of business conditions that can then give people the total picture. It really doesn’t matter where a company is today, nor does it matter where a company has been. All of that has already happened and cannot therefore be changed. What can be changed is the future and that’s where people need to concentrate their thoughts.

Think of a line chart as a beginning, a middle and an end. Historic values form the anchor upon which the line chart is constructed. The last current piece of data is the jumping off point and the extension of the line formed gives us an idea as to where our future “might” be.

Once we see the total picture, we can determine if it is heading in the right direction. Well, that’s part of the analysis, but not everything. The extended line chart gives us a hint of our future, but we still haven’t figured out if that’s where we want to be. You need a second line chart that acts as our target. Now we can see in an instant our past, one possible future and our target.

There are still two adjustments we might need to make. If the volatility of the data is significant, we may have to utilize some form of smoothing to make the data more understandable. Second, data does not follow a straight line path. Sometimes it’s increasing/decreasing over time. If that’s the case we may need to utilize some form of analysis that let’s us see these trends.

Now we can “see” the data and make a decision. If the trend seems to be within the acceptable range or budget we created, then no decision is required. If the trend seems to be heading in a negative direction, then we know we need to take action. No data analysis will ever tell us what to do, but this approach will help us determine if we need to do something and most importantly if the decisions we have taken seem to be having a positive affect.

One last thought. The key to the charts we have been discussing is that they give us an opportunity to become more proactive. If you see that your actual results are starting to trend in a negative direction, you can proactively take steps to make changes before the raw data indicates that you need to do something. Of course this is just another form of forecasting, but in this case the forecast is operationally oriented.

What information needs to be extracted and displayed?

While it is certainly easy to create graphs once the data values have been identified, the hard part is determining what you need to track. Remember, you can create any number of fields than can be used to feed your data analysis engine. Picking the right fields is the tricky part.

Think of an Income Statement. It’s got lots of data that could be graphed. Now think of virtually every accounting and ERP system’s ability to support drill down. If you can drill down from a data value to underlying information, then the information is of no use to you because it is being influenced by other data. Since you cannot track everything, you need to identify those basic factors that have the most influence on your business. Identify your profit drivers.

Maybe inventory turns in a distribution environment can be thought of as a Profit Driver. Don’t forget though that it’s not going to be possible to track every single item your carry. Maybe start with inventory turns as a whole, then by product line or possibly region. Where you start is not as important as your ability to quickly see where you may have a problem developing. Then you can drill down to a more detailed analysis.

Profit Drivers are a bit like the concept of a common denominator. You need to identify those values that are not influenced by other factors. If you can track and control these key business drivers, the Income Statement will take care if itself.

Summary

Effective business decisions drive business profitability. These decisions need to be rooted in facts that can be brought to light instantaneously. People do not have the time to guess. They need to know where to place their attention. They also need to know whether the decisions they make are having the desired effect. If there is too much data, the issues may remain clouded. If there is no way compare actual results against targets, how can you ever know if you are where you want to be or need to be. Finally, people need to identify those Profit Drivers that have the most significant impact on the organization.

Business Intelligence systems are the vehicles by which critical information can be extracted from an accounting or ERP system and used to not only monitor internal processes, but also support the strategic decision making process whereby firms can identify opportunities to create tactical advantages or meet market challenges. Since changes in tactics might need to be implemented quickly, there is no time to waste. You do not have time to query your Business Intelligence system and you certainly need to avoid the creation of erroneous data analysis.

If you are going to rely so heavily on your Business Intelligence system to support the decision making process, your analysis needs to be proactive and it needs to be managed by people that are Business Intelligence experts, not casual users. If you can determine in advance what your competition might do, then you can also determine in advance what you might do. Rather than waiting (and doing nothing) until your competition strikes, you can and should be using your Business Intelligence system to support the planning process, not support the reaction process.

Introduction

If you have reached that point where a critical mass of your employees have decided that your current ERP solution might need to be replaced, what’s your next step? Many firms just launch a search, assuming that at some point they will find the best suited ERP solution. Well….there’s just a tiny flaw in this approach. How can you search for a new ERP solution if you don’t know what you are looking for? If you were purchasing a sophisticated piece of machinery, you would need to define precisely the capabilities this machine would have to possess. You certainly would not call vendors and ask them to tell you what you need, but that seems to be the case when it comes to purchasing a new ERP software solution.

Rather than just calling ERP vendors as your first step, you need to determine not only what you require in terms of functionality, but also the outcome you require. A complex production machine needs to be able to produce a specified number of parts in a specified time frame at a cost that does not exceed a specified cost (cost per piece produced). This same rigorous cost analysis needs to be applied to the purchase of a new ERP software solution. What do you require (improvements) and how much is it going to cost to achieve this level of improvement?

Put very briefly you need to justify the cost of your new ERP business management solution. It really doesn’t matter whether you are going to spend $10,000 or $10,000,000. You need to determine exactly what you need in terms of functionality, by how much this functionality will improve or increase your operating costs, how much these improvements will cost, and finally of course the net return for this anticipated investment.

The question now becomes simple. How are you going to “find” the monetary gains that will be required to justify what could prove to be a sizeable investment?

Reduce Days Sales Outstanding

There are two sources of low-hanging fruit; reducing AR Days Sales Outstanding is probably the easiest target. If you sell on credit, it is very likely that your customers are not paying you on time. You might offer Net 30 days, but your customers might be paying you on average in 60 days, perhaps even more. If you can reduce your AR by just 5 days (seems like a very achievable target), you can generate $13,700 in cash flow for every $1 million in revenue (based on a 365 day year). This translates into $137,000 for a $10 million firm. Unlike operational improvements that can be counted on every year, an improvement in AR generated a one-time cash flow; substantial to be sure, but just once. Since a new ERP solution consists of a substantial one-time cost (purchase and implementation), this cash flow can be used to reduce your up-front investment.

While getting customers to pay you sooner, there are two additional opportunities for reducing AR. First, reduce the time it requires to actually generate an invoice. Time is quite literally money here and the sooner you submit an invoice to a customer the sooner you are going to be paid. Second, reduce invoice error rates. If you submit an incorrect invoice to a customer, it isn’t going to be paid until it’s correct.

Improve Inventory Management

The other substantial asset that can drive cash flow is inventory. If you can manage your inventory more effectively, you will be able to improve inventory turns and therefore generate a one-time cash flow just as I have suggested for Accounts Receivable.

Balancing inventory levels is a tricky business. While you can certainly reduce quantities on hand, this may lead to stock outs. You cannot sell what you do not have. In a manufacturing environment stock outs can bring the manufacturing process downstream to a halt. That’s why many large ERP solutions include sophisticated advanced planning applications. That’s also why lean manufacturing is becoming so important. Rather than having sufficient quantities on hand to avoid stock outs at all point in the manufacturing process, lean manufacturing extends the planning process out to suppliers as well as in-transit inventory.

Reduce the Cost of Purchasing

It’s amazing sometimes how expensive it is to purchase inventory. Take a close look at how many people are involved in the purchase process from initial vendor relationships to the point where purchased goods are received. Each time someone has to “touch” this process adds cost to the material being purchased. Given the sophistication of today’s ERP solutions, it should be possible to construct a process workflow that is based on the assumption that everything’s OK. Spend time (and therefore money) building a rock solid relationship with vendors whereby unit costs and delivery requirements are known and acceptable. Once the replenishment system has triggered a purchase notice, the resulting purchase order sets the best quantity (based on current forecasts and quantify price breaks). If the suggested order quantity falls within specified limits, the purchase order is sent to the vendor automatically with no touches from anyone in the purchasing department.

Once a purchase order has been issued, no human needs to become involved, unless the vendor submits a warning, or no advanced ship notice has been received. Only at that point should someone become involved. No reports need to be generated and reviewed at any point in the process.

Exception management should rule the purchasing process. The entire process should be based on the fact that a purchase order will be completed with no issues. If an issue is identified, users should be able to resolve the issue using the equivalent of an internal task management system.

Increase the Efficiency and Effectiveness of the Reporting System

Eliminate all printed reports, particularly those containing rows and columns. This is a soap box that I have jumped on for many years. My point is simple. Static reports are no more than a single picture in time. They give you no direction with respect to where conditions have been or where they might go in the future. Again I come back to the concept of exception management. Users need to be spending their time improving those critical operations that need to be improved. They don’t need to be wasting their time trying to figure out what needs to be improved.

Users waste precious (costly) time drilling down from static reports to underlying source data. Why not just start with the source data? Identify those lowest common denominator values that represent critical operations. It could be inventory turns or any other factor that represents those things that a company needs to do very well. If you need to drill down from what you consider to be a critical value, then you haven’t gone deep enough. Assuming that you correctly identify these critical values and concentrate on improving them, everything else in an income statement will take case of itself. Basically you don’t need an income statement except for those activities by a CFO that requires an income statement and balance sheet.

Having defined these critical values, track them over time (month to month or some other appropriate time frame). Use a line graph to track these values and a second line graph to specify where the values should be. Now you have a complete time-phased picture that you can use to identify where you need to spend your time. If the graph indicates no changes, a quick view is all you need and then you can go on to the next graph. I think a huge amount of time can be saved by not analyzing information that does not need to be analyzed. Conversely, you can spend more time improving those areas of the business that really need your attention.

Summary

Since this is a blog and not a formal article, I have but skimmed the surface. If you are contemplating the purchase of a new ERP solution, you have to specify precisely how you are going to be able to improve your business. Each of these improvements needs to have a monetary value attached to them. Your task at this point in time is defining a realistic transformation. What do you really want? How can you improve your business? What cost saving or revenue improvement can be associated with each improvement? If you don’t define what’s realistically possible, how can you possibly justify your investment in any ERP solution?

Business Reporting has always been one of the many soapboxes I love to jump on (http://www.accountinglibrary.com/blog/category/business-intelligence/, http://www.accountinglibrary.com/blog/category/business-alerts/, http://www.accountinglibrary.com/blog/category/exception-management/). While there is no doubt that sound business decisions are based primarily on the person making the decision, these decisions need to be based on sound information. Seat-of-the-pants decisions might work for small firms where the experience of the decision maker is the key element, but there is just too much information spread over too many decision makers for this practice to work in larger firms. In this case accurate and timely information must serve as the foundation upon which sound decisions can be based.

Why do you need this information?

Accounting and ERP systems can store and regurgitate just about anything you want. Actually it’s possible that for every transaction posted there can be 20, 30 or 40 fields that will allow you to slice and dice data any number of ways. The fact that you can carry reporting (intelligence?) to this level of detail doesn’t necessarily mean you should do it.

Start by asking yourself why you need the information. If it doesn’t lead to a decision, then you don’t need it.

In what format should this information be presented?

With the exception of audit trail reports, rows and columns are out. In no case can they lead to a decision.

Pie charts and bar chart are out as well. They, like row/column reports, are no more than a snapshot of a specific condition at a specific instant in time. People should never make snap decisions and that’s all this information can support.

Having said this, there is one use to which some specific data can be applied. Exception Management or Business Alerts are usually triggered by a specific value, but only if that condition falls outside an expected range. As an example, if a customer’s account exceeds its credit limit or a specific invoice becomes “x” days overdue, that condition should be brought to the attention of a named individual. Business Alerts handle the notification process while Exception Management gives users the ability to deal with an alert in a contact manager like application.

If rows and columns, bar charts and pie graphs are not acceptable, what’s left? Line charts are the only form of reporting that actually lets users develop a sense of the history and future of business conditions that can then give people the total picture. It really doesn’t matter where a company is today, nor does it matter where a company has been. All of that has already happened and cannot therefore be changed. What can be changed is the future and that’s where people need to concentrate their thoughts.

Think of a line chart as a beginning, a middle and an end. Historic values form the anchor upon which the line chart is constructed. The last current piece of data (the most recent value) is the jumping off point and the extension of the line formed gives us an idea as to where our future “might” be.

Once we see the total picture, we can determine if it’s heading in the right direction. Well, that’s part of the analysis, but not everything. The extended line chart gives us a hint of our future, but we still haven’t figured out if that’s where we want to be. You need a second line chart that acts as our target. Now we can see in an instant our past, one possible future and our target.

There are still two adjustments we might need to make. If the volatility of the data is significant, we may have to utilize some form of smoothing to make the data more understandable. Second, data does not follow a straight line path. Sometimes it’s increasing/decreasing over time. If that’s the case we may need to utilize some form of analysis that let’s us see these trends.

Now we can “see” the data and make a decision. If the trend seems to be within the acceptable range or budget we created, then no decision is required. If the trend seems to be heading in a negative direction, then we know we need to take action. No data analysis will ever tell us what to do, but this approach will help us determine if we need to do something and most importantly if the decisions we have taken seem to be having a positive affect.

What information needs to be extracted and displayed?

While it’s certainly easy to create graphs once the data values have been identified, the hard part is determining what you need to track. Remember, you can create any number of fields than can be used to feed your data analysis engine. Picking the right fields is the tricky part.

Think of an Income Statement. It’s got lots of data that could be graphed. Now think of virtually every accounting and ERP system’s ability to support drill down. If you can drill down from a data value to underlying information, then the information is of no use to you because it is being influenced by other data. Since you cannot track everything, you need to identify those basic factors that have the most influence on your business. Identify your Profit Drivers.

Maybe inventory turns in a distribution environment can be thought of as a Profit Driver. Don’t forget though that it’s not going to be possible to track every single item your carry. Maybe start with inventory turns as a whole, then by product line or possibly region. Where you start is not as important as your ability to quickly see where you may have a problem developing. Then you can drill down to a more detailed analysis.

Summary

Efective business decisions drive business profitability. These decisions need to be rooted in facts that can be brought to light instantaneously. People do not have the time to guess. They need to know where to place their attention. They also need to know whether the decisions they make are having the desired effect. If there is too much data, the issues may remain clouded. If there is no way to compare actual results against targets, how can you ever know if you are where you want to be or need to be. Finally, people need to identify those Profit Drivers that have the most significant impact on their organization or on their specific area of responsibility.

Forget about columns and rows. Forget about bar charts and pie charts. Adopt a proactive system that helps you track not just where you have been or where you are today, but where you could be tomorrow.

Introduction

One of the keys to increasing business profitability is giving each employee specific information they need to make sound decisions, the ability to concentrate on the areas of the business that require their immediate attention and an environment that allows them to collaborate with others to address issues that span more then their individual areas of responsibility.  Binding all of these success factors is one key concept.  If any business wants to achieve true success, they must be proactive, not reactive.  This article will discuss how people can use Business Metrics to identify critical issues before they can have a significant negative impact on the organization.

Background

Most classic business management reports are at their core no more than a picture of an organization at a specific point in time.  As such these status reports do not help people make decisions and do not effectively contribute to income and profit growth.

While this type of reporting was acceptable in the past (particularly since it was all we really had), we now have the technical ability to measure even the smallest events that drive any business.  The question is how can we build a framework for success?

Proactive Business Metrics

I cringe every time I see bar charts and pie charts as representative of what information users can extract from an ERP system and display on their “dashboard”.  While these charts certainly look pleasant (particularly when displayed in color), in the end they are nothing more than status reports and as such do not (or should not) drive business decision making.  Why? Bar charts and pie charts do not (and never will) lead to any form of decision or action.  A pie chart that displays the percentage of revenue generated by a firm’s top ten customers tells you nothing.  There is no basis for comparison to tell whether this is good or bad.  It looks pretty, but it leads to nothing.

Truly effective business metrics give you a basis for comparison that either tells you where you are going and whether you are above or below expectations. 

Let’s use Inventory Turns as an example.  Measure Inventory Turns each month and plot the values in a line graph showing the value for the past 12 months (or any time frame that is sufficiently long that a pattern can be discerned.  The user can visualize immediately where Inventory Turns has been and where it might be going.  It’s a good start, but not perfect.

Now add a system generate trend line that is calculated from the raw data.  In many cases the individual data points are so chaotic that it is difficult to see any trend, but the trend is what’s critical, not the individual value each month.  The trend line should not be a straight line, but a curve that shows whether Inventory Turns is improving or not.  A manager viewing this chart might be able to see instantly that the trend is acceptable and nothing needs to be done.  Next chart please!

Now let’s take this analysis one step further.  While the trend line in the example above may seem to indicate that Inventory Turns is acceptable, you need to add one final element.  How does Inventory Turns compare against expectations (budget)? Actually you could choose to not display the raw data as the trend line and the budget line are the two critical elements.  Now you have the complete picture: the actual trend and the budgeted trend.

This type of charting is used in the securities industry every day.  Computers track a stock’s value each day.  They also calculate a confidence interval above and below the trend line.  If the trend breaks above the upper confidence interval, a buy signal is generated.  Similarly, if the trend line breaks below the confidence interval, a sell signal is generated.  That is precisely what could be done to assist manager track inventory turns or any other defined success factor.

Business Metrics Delivery

Assuming that you have identified the Business Metrics that drive your business, how do you effectively deliver this information to named users?

1. Identify specific Business Metrics that drive an organization’s success and that lend themselves to a budget/actual comparison. 
2. Give users the ability to display any chart full-screen.
3. Give users the ability to drill down to the independent variables that drive the data displayed.  As an example, Inventory Turns is driven by sales, purchasing activities and receipts.
4. Give users the ability to view the actual data in table form.
5. Do not display information that is of no concern.  If the values in the chart are acceptable, give users the ability to either not have these charts displayed or display them at the bottom of their dashboard.
6. Give users the ability to request that a chart be included in their display list the following month.  In this example, Inventory Turns might fall within an acceptable range next month, but the user may want to “see” for themselves.
7. Utilize a red light / green light summary whereby users can view a list of their personal Business Metrics.  Actually all you would have to display is the name of the metric and a red/green light indicator.  This would be particularly useful if a person was tracking a significant number of charts.  If the information is acceptable, the green light will be displayed.  Obviously if the information is not acceptable, the red light would be displayed.  The user can then hyperlink to the metric of greatest interest.
8. Give users the ability to view the charts of those people reporting to him/her.  While we should assume that people will be attentive to their areas of responsibility, some managers might want to review the status of information tracked by people on their team.  Alternately, the system could be set up to track individual metrics as well as team metrics.

Summary

The real key to business success isn’t trend line charts.  It’s the recognition and subsequent tracking of what’s important to the business.  Before you can succeed in business, you have to understand what drives your business.  Then you concentrate on these basic or lowest common denominator success factors.  If you get them right, then your Income Statement will take care if itself.