Tölvumál - 01.10.2013, Blaðsíða 28

28
If you are not monitoring the health of your company’s finance,
customers and time, you are doomed to failure. A dashboard and solid
metrics are the lifeblood of any company. A well designed and
implemented dashboard can be the difference of being successful or
shutting down. Understanding what is needed to create worthwhile
metrics that are useful to the whole organization and displaying them in
a way that they can’t be misinterpreted isn’t an easy task.
Does your business have a glance able tool for you to make decisions?
Is there a single place where you can go to see all the important key
performance indicators? Can you easily take the information with you
on the go and access it from any device anywhere? It is unlikely that
you said yes to all three questions. Having a single place, accessible
from anywhere which quickly conveys meaningful information is a
difficult task. It’s not an all or nothing proposition, having metrics is still
better than having nothing. Having a dashboard only available at work
is also better than no dashboard at all.
So how do we go about creating metrics to track the business and a
dashboard to monitor the health and progress of your business? Before
we get into design and implementation, we need to examine your
company culture to find out what metrics you already have, which need
to be tracked and how to get into a different mind-set about quality,
innovation and leadership.
# meTricS
Business metrics tend to come in two flavours; vanity metrics and
actionable metrics. While vanity metrics are nice to show-off, they don’t
mean much to the company. For instance, the number „hits“ on a
website is purely a vanity metric. Every „hit“ to your website means
something different. Maybe it’s the same user coming back lots of time,
maybe it is lots of users visiting only once, maybe your website has lots
of images and each counts as a „hit“ when viewed. Generating more
„hits“ isn’t solving a problem, achieving a goal or even comparable and
therefore it is a vanity metric.
Instead of looking at the number of „hits“ a website gets, why not look
at the number of sales. Now we are moving away from vanity and
beginning to look at actionable items. Rather than increase “hits“ we
want to increase sales to generate more money. We can get more
granular and create a „sales per customer“ metric. There is a direct
impact and feedback loop for actionable metrics. You track issues and
goals, fix or change them, and check the progress. Measuring the right
things are more important than just measuring to be measuring.
Edward Deming (deming.org) was a 1950s statistician and a professor
who turned around the Japanese industrial system after world war II.
Many of his ideas and teachings are still in place today. There are many
books on the topic but it boils down to his famous 14 key points. One
of the points highlights the need to remove quotas. If an employee gets
paid based on the number of units shipped, he argues this is bad
practice. Quotas are easy metrics to track and put on a dashboard. If
you are building 10% more units this quarter than last quarter it makes
sense to have this metric on your dashboard. His argument isn’t about
the quantity of units, but about the quality. The real actionable metric
should be: What percentage of units shipped these quarters are
defective? Quality is the real goal - not shipping more units, but to
shipping fewer defective units. The same will be true for increasing
‘sales per customer’. We run the risk of either tricking customers into
buying more stuff they don’t need or charging more to inflate our sales
numbers in the short-term. The problem will be that those customers
never return and we lose out in the longer term. Maybe a better
customer metric would be the number of repeated sales or customer
loyalty? All these issues and caveats makes creating a good metric
much more difficult than it might seem on the surface.
Once you better understand what metrics your company relies on, you
realize that each department has different metrics they need to track.
Accounting for instance, needs to know about the number of defective
units, repair and return costs, but it isn’t something they can directly fix.
On the flip-side, knowing the average time a late invoice is paid is
obviously more important to the accounting team than it is to other
parts of the workforce. Just because a metric isn’t useful for your team
doesn’t mean that it should be ignored or not tracked. Everyone needs
to work together to make improvements.
Once you have identified your metrics, it is time to look at how to
convey those raw numbers and trends in a way that anyone can
understand. This is where we get into dashboard design. Remember
that even the most beautifully design dashboards are only as strong as
the metrics they show.
# daSHboardS
Let’s discuss some of the early stage questions you should be asking
yourself and the team as you create a centralized place for you to keep
an eye on your business.
First, we need to decide if we are going to build our own dashboard or
buy one. A dashboard can be a stand-alone application, but it can also
simply be a webpage. Both of which are available to buy or build. Since
most of the data you’ll be accessing will be across the network or even
the internet, it makes sense to also use web-based technologies to pull
together the design of your dashboard. Using an app might be cleaner
and more tightly coupled to the operating system, but it also limits you
specific devices and isn’t always as flexible, adaptable or as quick to
change as a web-based solution.
Designing a good dashboard requires a lot of thought and planning.
Not every metric you want to present will come as a cell in a spread
sheet or from an API call. There will be integration issues across
departments and legacy systems. If coordinating or getting „buy-in„
from management is too difficult, then maybe not everything needs to
meTricS and
daSHboardS
Brian Suda, owner of Skolapulisnn.is and developer