Skip to content

Nomencalature in reporting – Build it for the ‘next’ guy

Truth.  Sometimes, other people tell us what to do.   Anyone in BI, data visualization, analytics, reporting, etc  can tell you this – we’ve all been there.     We generally don’t do this for ourselves and others write our paychecks.   Now that we’ve gotten that out of the way…

The most important piece of any report is the text.   Field names,  column headers,  labels – the glue that informs the viewer what he/she is looking.

Today,  I watched as a developer gave a requester what he asked for – a column name labled ‘MoM’ for “Month over month”.

Here’s how the conversation went:

Me:  What does “MoM” stand for?

Developer: “Month over month”

Me:  “Which month over which month?”

Developer: “This month vs last month?”

Me:  “Why not call it current vs prior then?”

Developer:  “This is what the user asked for, and he knows what it means…”

Answer:

This person will move on, take a vacation, get promoted, be out sick,  you don’t write a report for the person who requests it – you write it for the next guy.

Measuring Brand Loyalty – Why ‘new vs returning’, Visits/Unique Visitor, and Return Visits in General are all CRAP!

Most major sites have different ways of driving traffic. For example, daily newsletters, paid campaigns such as AdWords, and search engine optimization. At the overall site level, the ‘new vs returning’ metric is generally much more reflective of the marketing mix, then of any particular campaign, content, site improvement, or brand loyalty in general for that matter.

For example, in the scenario I describe, and improvement in SEO rankings would shift the mix toward more new visitors. Weakened rankings, higher return visit % as a larger cut of the audience are the registered daily newsletter folks.

Attempting to segment new vs returning by traffic source doesn’t cut it either, as these just become proxy for performance of the source, and not a particularly good one.

The best measure I’ve been able to come up with for brand loyalty is simply “return visits to the site homepage”. This is agnostic of how google ranks you, social and paid traffic drivers, etc, and gets to the point – how many folks are typing in your URL directly.

What do you think?

Does BI belong in IT?

Sounds crazy, but after 20 some-odd years of doing BI in an IT world, I’ve spent the past few doing BI as part of “the business”, and I can’t tell you how refreshing it’s been!!

Ok – well, not exactly ‘the business’; more like a separate business unit, complete with a Chief Data Officer who reports to the CEO. IT in our organization is the realm of apps-dev and tech support.

Oh, and I get it, BI practically has “information” and “technology” in it’s name – this guy’s nuts, right? But when you look at the typical IT organization, which tends to take it’s cues from an application development mindset – maybe not such a perfect fit.

Having the expertise and mandate to extract/present meaningful insights from data along with the know-how to work it has it’s advantages.

Curious to hear other’s perspectives!

What’s your BI Mission?

If you’re reading this – chances are you work in Business Intelligence and love what you do.

From time to time, however, we need to take a good look at ourselves and ask ourselves the bigger, more existential question like “why am I here?”, or more importantly, “why is someone paying me to be here?”. The answers are not always as easy and as obvious as they might seem.

A mission statement is a great tool for this. It forces you to really think about what you’re doing and why you’re doing it. It also helps align your team, getting you all rowing in the same direction, and comes in handy big-time when you need to present your wares to a larger audience.

Here’s what I came up with for my current place of employment.

“Maximize the use and function of the company’s data assets to help optimize the performance of the enterprise though improved decision making and planning.”

Anyone got a better one?