Facts about Perkins Consulting
- Focused on turning data into actionable business information.
- Founded in June 1994
- Based in the PacWest Center in downtown Portland, Oregon USA
What we Do
It's about people. In the end, every successful business decision comes down to making the right choices about people--customers and employees.
And those right choices can only be made by organizations that thoroughly understand who their best customers and employees are and what they value.
At Perkins Consulting we use information to give our clients a complete understanding of who their best customers and employees are
and what motivates them. We work closely with our clients to integrate this knowledge into their unique business strategy and then we design processes
and technology to make it actionable. Whether its someone on the executive team or on the front line, we make sure that each person has exactly
the knowledge they need to make the best decision.
- Right Information
- Right People
- Right Actions
- Right Outcomes
We've specialized in providing information solutions to mid-market companies in the northwest for over 15 years. We know exactly what
kind of information is most valuable to these organizations; our projects deliver increased margins, reduced turnover,
decreased working capital and new revenue.
We make information powerful by integrating, enhancing and delivering data so that it becomes actionable information.
- Integrating data from disparate sources so that the result is greater than the sum of the parts
- Enhancing data to make it focused and usable information
- Distributing information to the right people at the right time
Integrating
There are many different ways to categorize data, but some of the more common ones are:
- Transactional data (e.g. a customer and salesperson linked through a sales
transaction)
- Behavioral data (e.g. how customers navigated a web site)
- Attitudinal data (e.g. results from an employee survey)
Organizations collect volumes of highly useful data in these different categories, but since the data is housed in different source systems,
such as ERP systems (Enterprise Resource Planning) and CRM (Customer relationship management) systems, they need to integrate
the data in order to better understand the informational relationships that matter.
One of our customers, for example, wanted to see how regional sales were impacted by different sales incentives. The data for
regional sales were housed in an ERP system and the data for promotional incentives were housed in
a specialized hosted application. We integrated the
data from these two different systems by creating a data mart. Then we used business intelligence tools that allowed the client to
see the correlation of how sales in each region were impacted by promotional incentives.
Combining data from different systems and sources into a data mart allows our clients to look at their business across functions and processes.
Instead of trying to glue together bits and pieces of data from various reports to form a collage of guestimates, decision makers are able to get the
complete picture of how their organization is doing.
Enhancing
Data is enhanced in several ways.
- By adding context or perspective that didn't exist in the source systems. For example, we may add new summarization categories
to data from a labor time collection system so that reports and analyses can easily have groups of related activities rather than
being confronted by a very long list of codes.
- Creating simple and easy to use custom applications to add descriptive attributes (e.g. explaining what certain codes mean) and
hierarchial structure (e.g. defining product lines to hold products for easier grouped reporting.) Our team has built such applications
as both web and Windows applications.
- Extending the clarity of data by providing descriptions for data that may only be expressed as codes in the source systems.
This allows reports to include things like a status description (e.g. Open, Closed) rather than just cryptic codes like X or O.
- Eliminating the risks of double-counting by using dimensional data modeling to flatten many to one relationships for easier
and more accurate reporting. Rather than having to use complex reporting tricks to summarize any header detail relationship (e.g. an
Invoice and its invoice lines) the dimensional structure flattens this out and makes it easy to either sum up the details or report
on a total already contained in the header.
- In the dimensional data mart we can also create valuable objects to make reporting much more powerful. One example of this is
the use of snapshot tables to capture the state of transactions over time. Common uses of this technique are any kind of long-lived
transaction such as an insurance claim.
We also enhance information by capturing budget and forecast information using modern web-based applications rather than
complex Excel templates. Having budgets, forecasts and targets to compare against actual performance puts organizations on the
path to
Performance Management
Distributing
Distributing data efficently is about much more than just having a lot of reports. The business value comes in delivering the right
information to the right people in the right format. We've selected and worked with many different platforms for
business intelligence and have learned a lot about the key requirements. We've
also been involved in building custom platforms that leverage off the shelf or open source components.
All of this experience has helped us constantly refine what we believe to be the essential components of a successful BI platform
for distributing information.
- The platform should support specialized skills. The organization shouldn't be dependent on a small number of geniuses to understand
the data and handcraft reports. The platform should allow for separate people to effectively combine their specific
skills in understanding data, understanding business strategy, creating reports and managing users and security.
- Any reasonable platform for distributing information should include all of the components of a good security system including
authentication (making sure we know who is using the information), authorization (only allowing users to see what they should be allowed
to see) and auditing (keeping track of who has accessed what.)
- The management of metadata should be a fundamental part of the platform. This metadata includes valuable information like: where did this
information come from in the source systems? How was it derived? How up to date (recent) is it?
- The days when business intelligence platforms supported only one delivery format are fortunately past. Any reasonable platform supports
delivering information via the web, Excel, XML and CSV.