This is not just a single year phenomenon but the result of strong upward trends in IT investment by the Information Services industry since at least 2000.

The information services industry

The industry classification system used by the US government largely changed in 1997.  As a result of NAFTA, Canada, Mexico and the US collaborated on a new classification system called the North American Industrial Classification System (NAICS) (the US was previously using the Standard Industrial Classification (SIC) system).   Many sub-industries were reorganized in the new system.  And a new “industry” was created representing sub-industries involved in some way with providing  “information” services.

Although there are officially six, 3-digit NAICS Information Services sub-industries, the US Bureau of Economic Analysis (BEA) collapses these into four:  Broadcasting & Telecommunications, Information & Data Processing, Motion Picture & Sound Recording and Publishing.  Since our investment data come from the BEA, we will focus on these four sub-industries here.

Software investment

One of the IT investment themes in recent years has been the transition to “cloud-based” solutions.  In the case of software, Software as a Service (SaaS) is supplanting software installed on local computers (aka “packaged software”).   The BEA does not break out SaaS spending from other types of software investment.  So, unfortunately, we cannot tell whether SaaS is simply cannibalizing packaged software or whether it is contributing to net growth.

However, we can see that there has been strong growth in software investment in the Information Services industry since 2000.  As the interactive dashboard below shows, software investment grew by 9.8% per year per worker (measured in 2009 $) between 2000 and 2015, the most current year for which BEA data are available.  This is higher than the national average rate of growth in software investment of 7.9%.

As you read our analysis below, we invite you to follow along by making changes to the dashboard to show the trends we are discussing.  Changes are made in the “filters“ “Select industry”, “Select years” and “Select investment type” found at the top of the dashboard.  Also selecting “Select sub-industry to highlight” will bring the selected sub-industries to the forefront in the bottom two line charts.

There is wide variation, however, within the Information Services industry.  Software investment by the Information & Data Processing sub-industry has been growing by a whopping 15.9% per year per worker since 2000.  However, this has moderated some in recent years (8.0% per year over the 2010 to 2015 period).  In 2015, the Information & Data Processing sub-industry invested over $65,000 per worker in software (2009 $).

At the other end of the scale, Broadcasting & Telecommunications has been investing at only a 3.5% annual rate since 2000 (5.1% since 2010).  In 2015, this sub-industry was investing over $12,000 per worker in software.

Storage investment

Another but related IT investment theme has been the transition away from physical storage to virtual or cloud-based storage…think DropBox.  As we saw in a previous article, according to BEA data, total investment in physical storage (in 2009 $) began to decline in the US private sector in 2000.

However, not all industries (and sub-industries) have shown the same effect.  In the Information Services industry, physical storage investment per worker has declined by 1.4% per year since 2000 (3.8% since 2010).  Use the “Select investment type” filter to affect this change in the dashboard.

The Information & Data Processing sub-industry continued to invest in physical storage at a 6.7% annual rate since 2000.  It was the disinvestment by the Broadcasting & Telecommunications sub-industry (-7.3% per year) that is reflected in the overall Information Services industry negative growth of 1.4% per year.

However, in recent years, since 2010, the annual rate of change in storage investment has been negative across the entire Information Services industry, with Motion Picture & Sound Recording sub-industry leading the charge at -19.1% per year.

Other stories

There are many more stories the data can tell.  And we invite you to play with this dashboard to uncover them.

For example, a previous article discussed investment trends in the Healthcare industry.  Does using investment per worker (the above dashboard) change any of our conclusions in that article?

Drop us a line and let us know what you find.

The post Information Technology Investment in the US Information Services Industry appeared first on KDD Analytics.

Interactive investment dashboard

The interactive dashboard below shows IT investment for a group (1 or more) of user-selected, 3-digit NAICS industries (blue).  The rest of the US market is shown in gray.  The default selections are the 3 healthcare sub-industries of Hospitals, Ambulatory Care and Nursing Facilities (see “Select industry” in the dashboard).  The top chart shows investment over time in 2009 $; the bottom chart shows CAGR over time.

As you read our analysis below, we invite you to follow along by making changes to the dashboard to show the trends we are discussing.  Changes are made in the “filters” “Select industry”, “Select years” and “Select investment type” found at the top of the dashboard.

Healthcare information technology investment

In 2015 (the last year for which BEA data are available), the Healthcare industry invested $10.6 billion (2009 $) in computers, software and communications equipment (e.g. phone systems, internet and networking equipment).  62% of this was invested by Hospitals, 32% by Ambulatory Care providers and 6% by Nursing Facilities.

Note that Healthcare industry spending on information technology consists of more than equipment and software.  A sizable share is spent on IT services (such as system integration and consulting) and IT outsourcing.  Forrester estimates that the Healthcare industry spent over 3.5 times what it did on equipment on IT services and outsourcing in 2016.  Over 70% of this additional spending is on outsourcing alone.  The BEA data only cover “fixed asset gross investment”.

Of the three “equipment” components of Healthcare information technology investment, software has exhibited the strongest growth (see “Select investment type” in the dashboard).  Since 2000, software investment has grown by a compound annual rate of 7.6%, mirroring the growth seen by rest of the US market (7.9%).

Pop quiz: Using the dashboard above, of the three healthcare sub-industries, which one had the highest rate of growth in software investment over the 2000 to 2015 period? It may not be what you first suspect.

As of 2015, Healthcare investment in communications equipment has grown by 4.0% per year while computer (hardware) investment has grown by 3.6% per year.  Nursing Facilities exhibited the strongest growth in hardware investment at 5.7% per year.

Note that this is total investment which can rise due to higher demand at each facility and/or an increase in the number of facilities over time.

Computer hardware investment

Digging deeper into computer (hardware) equipment investment reveals some startling trends.

Both PC and server investment have been growing at 6.7% and 4.3% per year since 2000.  However, the annual rate of change in PC investment has been negative since about 2007.

Also, the annual rate of change in physical server investment has been essentially flat since about 2005.  This latter trend is consistent with movement towards virtual and cloud-based servers.

The real standout is investment in physical storage equipment.

Physical storage investment by the Healthcare industry has actually declined by 1.6% per year since 2000.  The most likely reason for this trend is again the movement towards “cloud” storage solutions.  Physical storage investment peaked in 2007 in the Healthcare industry.  Selecting “Select years” in the dashboard to show “1980 to 2015” and the “Select industry” to display to “All” reveals that the peak in physical storage equipment investment in the total, private US market occurred 7 years earlier in 2000.

Printers have also apparently lost their luster.  Healthcare investment in printers has declined by an annual rate of 1.7% since 2000.  Again, looking over the entire 1980 to 2015 time frame, printer investment peaked in 2002 and has been in a downward trend every since.

Healthcare investment in displays increased by 3.0% per year since 2000.  However, since 2004, the trend in annual investment has been flat.  Nursing Facilities exhibited the strongest growth in display investment at 5.3% per year over the 2000 to 2015 period.

What have been the IT investment trends in other industries?

We invite you to play with this dashboard to conduct a similar analysis for any one (or multiples) of 60+ industries.  The limitation, as alluded to above, is that the investment shown is total or aggregate investment.  On a per business or employee basis, trends may be quite different.  We will pick this up in a future article.

The post Healthcare Information Technology Investment Trends appeared first on KDD Analytics.

Why does information technology investment per worker matter?

While information technology (IT) investment per worker has grown over time across all industries, there clearly are industries that use IT more intensively in their production of goods and services.  Knowing how this intensity varies across industries can inform public policy decisions.  For example, the demand for labor may be impacted by the amount of IT workers have to work with.  Thus any policy (e.g. tax) targeting business investment decisions can have a differential effect on labor markets depending on the intensity with which IT is used.

Knowing how IT intensity varies across industries is also useful from a micro perspective.  How much ABC Corp located at 100 Main Street, Any Town USA spends on IT is probably unknown to B2B marketers.  However, knowing in which industry ABC Corp plays and how many employees work at the 100 Main Street location can yield a ball park estimate of how much ABC Corp can potentially spend on IT.

Such information would be quite useful for IT vendors selling into ABC Corp.

Where does one find data on IT intensity?

The big IT market research houses, such as Forrester and IDC, provide industry-level estimates of IT spending.  But usually at an aggregated industry-level (e.g. 2-digit NAICS or 1-digit SIC).  And they are unlikely to provide employee counts.

Fortunately, the US Bureau of Economic Analysis (BEA) provides estimates of investment in IT “equipment” (computers, software and communications).  While they do not provide estimates of spending on IT services (e.g. IT consulting, system integration and outsourcing), the BEA data provide insights into how much businesses are spending on the three other main components of IT.

Moreover, the BEA provides these data at the sub-industry, 3-digit NAICS level; they provide estimates of employment at this same level of disaggregation; and they provide these data every year.

What more could a data junkie want?

Information technology investment per worker (IT intensity)

IT intensity is defined as gross investment in hardware, software and communication equipment per worker.  In 2015 (the last year for which data are available), IT intensity averaged about $3,100 per US private sector worker (in 2009 $).  This is 3 times what it was in 1998.

However, this national average hides the dramatic variation across industries.  Below is an interactive dashboard which shows IT intensity expressed as an index (industry intensity/average intensity) for 2015.  Not surprisingly, the Information Services industry invested over 8 times the national average; the Agriculture industry just .03 times.

Drilling down into the Information Services industry (click on the bar chart) shows that it is the Data Processing and Broadcast and Telecommunications sub-industries that account for the lion’s share of Information Services industry investment (see the bottom chart).  Again, not too surprising.

The dashboard allows one to select the type of IT investment.  Selecting “Communications” reveals the intensity of investment in communications equipment (phone systems, internet and networking equipment).  Again, the Information Services industry leads the pack as a result of heavy investment in the Broadcast and Telecommunications sub-industries.

However, the bottom chart shows the ranking of all sub-industries (3-digit NAICS).  Interestingly, the Water Transport and Pipeline Transport industries are relatively intensive users of communications equipment.

If anyone knows why, drop us a line, we would like to know!

Other interesting findings…selecting “HW – Servers” reveals the intensity of investment in physical servers.  The top three sub-industries are Federal Reserve Banks, Rental and Leasing and Credit Intermediation.  At the 2-digit NAICS level, it is the Finance industry that is the top investor in servers, followed by the Wholesale industry.

We will leave it to the reader to play with the dashboard and uncover other interesting stories (e.g. in the Manufacturing industry, which sub-industry invests the most in storage equipment? in printers?).

Takeaway

Our takeaway is that although the “usual suspects” appear as high intensive users of IT, at the sub-industry (3-digit NAICS) level, there is wide and not always predictable variation.  Especially when drilling into the different categories of IT investment.

What is your takeaway?

In a future article, we will take up this subject again and examine in more detail how information technology investment per worker (IT intensity) has changed over time.

The post US Information Technology Investment per Worker appeared first on KDD Analytics.

A client’s customer was using an “off the shelf” application for prospect scoring.  This application used a very limited set of firmagraphics to construct its scores.

The customer knew our client had access to a wider array of business site data.  So they asked if we could measurably improve the performance of their prospect scores.

The customer used prospect scores to identify high potential prospects (business sites) in a information technology (IT), B2B marketing list.

The state of B2B business data

There are a number of vendors which provide data on the characteristics (firmagraphics) of business sites.  Dun and Bradstreet, InfoGroup, Compass, Orb, V12, to name a few, provide data such as number of employees, industry code and revenue.

Data vendors vary in terms of how many data fields they provide, in the amount of “white space” in these fields, as well as in how many business sites they cover.  And, of course, in price!

But all data vendors are constrained by the fact that, unlike consumer demographics, business firmagraphics tend to be more limited in number.

Another set of data vendors provide “enhanced” B2B data.  In the IT B2B marketing space, vendors like HG Data and Aberdeen provide data on the presence of certain technologies and technology vendors, counts of technology (e.g. PCs) and estimates of IT spend.

Enhanced B2B data coupled with base firmagraphics yields much more fertile ground for prospect scoring than just base firmagraphics alone.

What are B2B prospect scores?

A prospect score is a number assigned to a list of non-customers to be contacted in an outreach campaign .  After sorting this list by the score, the non-customers who are the “best” prospects will rise to the top.

These “best” or “high potential” prospects typically resemble current customers.   Converting these into actual customers is the job of the sales team.  The prospect score just tells them who they should focus on first.

Data analysts typically construct prospect scores through statistical modeling of a sample of customers and non-customers and their characteristics.

Can enhanced B2B data and methodology improve prospect score lift?

Since our client had access to enhanced B2B data along with base firmagraphics, we structured a test for our client’s customer as follows:

• First, construct a prospect score using only the few firmagraphics the customer was currently using;
• Second, construct a 2nd score that used these same firmagraphics plus enhanced B2B data fields such as IT spend, technology counts and indicators of technology presence;
• Third, construct a 3rd score using all available data and paying more careful attention to data preparation and modeling techniques.

To gauge model performance we used “lift”.  In short, lift is the degree to which a score identifies more potential customers compared to using some other score or no score at all.  We discuss lift in more detail in a prior article.

Findings – base score

Six (6) firmagraphic fields support the base score:  enterprise revenue and employees, site employees and type (HQ, branch, standalone), industry and metropolitan area.  The lift chart below shows that the resulting score does identify about 75% more customers (compared to using no score) in the top decile.

But the lift chart does not exhibit a smoothly declining lift from the highest to the lowest decile.  Nor is the lift decline very pronounced.  This suggests that the “discriminatory” ability of the score (i.e. the ability to tell the difference between a customer and a non-customer site) can be improved.

How do you construct a “lift chart”?  Start with a list of companies, some are customers and some are not.  Build a model and then score this list.  Sort the list by the prospect score, from highest to lowest potential.  And break the list into deciles.  Find the actual occurrence of customers in each decile and express this occurrence in terms of an index.  Index values > 1.0 indicate the score does a better job than using no score.

Findings – enhanced score

Various site-specific data fields for presence of technologies, IT spend and technology counts, as well as the base score’s firmagraphics, support the enhanced score.  The resulting lift chart is now smoother and exhibits a more pronounced decline in lift from the highest to the lowest decile.

That is, the discriminatory power of the score has markedly improved.  In this case, by about 15% in the top decile.

Findings – complete score

The final score uses a more sophisticated data preparation and modeling methodology.  And additional firmagraphics.  This adds an additional 5% to lift in the top decile.

The lift chart now suggests that, in the top decile, the score is 2x more likely to identify a business site that is similar to a current customer (compared to no score).

Findings – summary

Another way to look at this is to consider what happens over the first 30% of the file, not just the top decile:

• A prospect score using enhanced B2B data fields in addition to base firmagraphics identifies 12% more potential customers (than using base firmagraphics alone);
• Using a more robust methodology can boost this by another 6 percentage points.

Both data and methodology matter

The take away from this exercise?  Both data and methodology matter.  But methodology will only take you so far.

The greater the number of enhanced B2B data fields available, the more likely the discriminatory power of the model will improve.  The caveat, of course, is that these data fields need to be materially different from each other.  If they are all highly correlated with each other, then additional “similar” data fields won’t matter.

Of course, your mileage may vary in terms of the impact on lift.  But augmenting your marketing data with enhanced B2B data and using an appropriate methodology to construct your prospect scores is likely to yield a positive ROI.

The post Enhanced B2B Data Can Markedly Improve Prospect Scores appeared first on KDD Analytics.

Their goal was to determine how much “lift” the prospect score provided to their outreach campaign for their information technology networking solution.  That is, they wanted to know if the score led to (and by how much) the identification of more, higher likelihood prospects than using no score at all.

Their very thoughtful test yielded a result that surpassed even our, admittedly biased, expectations.

What are B2B prospect scores?

A prospect score is a number assigned to a list of non-customers to be contacted in an outreach campaign.  When this list is sorted by this score, the non-customers who are the “best” prospects rise to the top.

That is, if the prospect score works as expected.

These “best” or “high potential” prospects typically resemble current customers.   Converting these into actual customers is the job of the sales team.  The prospect score just tells them who they should focus on first.

What is lift?

“Lift” is marketing-speak for how much a B2B prospect score sorts a list of non-customers, from highest to lowest potential, compared to some other score, approach or no method at all.  Lift is presented as a percentage, such as 133%.

Consider a list of companies, some are customers, most are not.  Now sort this list by the prospect score, from highest to lowest potential.  And break the list into deciles.  A chart like this can then show the lift attributable to a prospect score built using this list.

This chart shows that over the first 30% of the sorted list (ordered by the prospect score), 70% of the actual customers can be identified (blue line).  This yields a lift attributable to the score, compared to using no score (a random selection in this case indicated by the red line), of 133% (i.e. [(70-30)/30]*100).

Of course, the lift that will actually occur in an outreach campaign targeting only non-customers will be different (and most likely lower).  But a prospect score that performs well in testing should also perform well in practice.

The size of the lift depends not only on how well crafted the score is but also on which alternative method the score is compared.  The lift attributable to one score vs. another might be very small.  However, compared to using no method for ranking non-customers, the lift could be quite large.

Why is lift important?

B2B prospect scores which yield appreciable lift reduce the cost of outreach campaigns.

Continuing with the above example, in order to identify 70% of actual customers without using a score, 70% of the marketing list would need to be used.  But by using the score to sort the list, these same 70% can be identified by using only 30% of the list.

This yields a cost reduction of 57% as the remaining 40% of the list does not need to be contacted (i.e. [(70-30)/70]*100).

So how do we build prospect scores?

Here at KDD Analytics, we build B2B prospect scores using statistical models.

We build these models using a sample of customers coupled with a randomly-selected sample of non-customers.  The goal is to arrive at a model that maximizes the differences (“discriminates”)  between customers and non-customers.

These differences are based on a set of “explanatory” characteristics.  In B2B, information technology marketing, these are foremost firmagraphics (such as number of employees, industry sector, enterprise revenue, etc.).  But they also include measures of information technology used at the business (such as spending on computers and software, likelihood of a certain vendor presence, number of PCs, etc).

Back to the test…

The customer of our client divided a prospecting list into two samples.  One sample used our prospect score, the other did not.

Then they initiated an actual calling campaign for their networking solution using both lists.  When the two campaigns were finished, our client’s customer compared the results.

In general, the degree of lift depends on the definition of a “successful” call.  For this test, a successful call was one in which the prospect expressed a genuine interest (follow-on call scheduled) and/or requested a quote.

The results of this test were a staggering 300%+ lift from using the KDD Analytics prospect score.

Granted this is a very high lift and we have to assume that the customer calculated lift correctly.  But it is consistent with our experience in B2B, information technology marketing.

B2B prospect scores can measurably improve outreach campaign performance.

Needless to say our client’s customer was extremely impressed with these results.  So much so that they immediately renewed their contract with our client for additional data sales and analytical services.

Of course, your mileage may vary.  But using B2B prospect scores to target potential new customers can measurably improve your marketing ROI.

The post B2B Prospect Scores – Improving Marketing ROI in the Information Technology Vertical appeared first on KDD Analytics.