« April 2008 »
S M T W T F S
1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30
You are not logged in. Log in
Entries by Topic
All topics  «
Ad review
Advertising
Marketing
Media
Online marketing
Blog Tools
Edit your Blog
Build a Blog
RSS Feed
View Profile
Stan's betaBlog: media marketing communications culture
Thursday, 3 April 2008
Tips from Forrester on getting comprehension -and buy in- on metrics from bosses
Topic: Online marketing

If you are having trouble getting your senior managers to understand and act on Web data, just “darn it.”

No, that’s not advocating muttering under your breath or even swearing out loud at them. Rather it’s the acronym from Forrester Research Web marketing manager  Corey Mathews for a simple strategy for data and analytic people to get the right information to decision makers in formats they can use: DARN IT - Define, Automate, Report, Nudge, Iterate.

“Your managers aren’t stupid, they’re just busy,” said Mathews during his presentation to the eMetrics Marketing Optimization Summit in Toronto at the beginning of April.

“We’re the experts in understanding the data. The issue is reporting to management in ways they can understand,” he says. “Executives need information in ways they understand and that they can make decisions on.”

Prior to presenting, Mathews informally surveyed analytics heads in 42 organizations around the world on their reporting methods and issues. Precisely what is reported, and the formats and frequency of reporting, may differ, he says. But the challenge of effectively sharing the right data with the right people is a “common frustration” in most organizations-even the enlightened Web-based businesses.

Information needs to be presented “carefully, repeatedly and patiently,” says Mathews, and always with an eye to what’s most important to managers and the organization-and in terms they are comfortable with. And that inevitably includes making sure you report on “the money,” especially ROI.

Define

Since what’s most important does and should vary from organization to organization, “defining” what success for the enterprise is and how those goals are measured is the first step in any communications strategy.

The primary data that is reported should speak directly to the organizations’ over-all business goals.

Automate

Metrics and research people often find themselves answering the same questions from different managers over and over, or being asked to do up special reports drilling into data that’s of special interest to one group of stakeholders, often on extremely short notice. And in every organization there are always a few managers who like to be able to dig into data themselves when they want it.

Where possible, automatically generated data streams and reports on FAQs or regularly asked questions should be created. This might be a dashboard with real time stats on sales or conversation rates, or it could be an e-mail with weekly or monthly updates on key stats.

This allows users to access data on their own, and frees up analytics people from constantly getting pulled off projects to do unscheduled updates and reports.

Report

It’s what it should be all about, but a remarkable number of organizations don’t have a thought-out plan for reporting Web data. Figure out a plan and schedule for who needs to see which reports and how often.

Not every report needs a formal face to face encounter, but there should be some of those planned, whether monthly, quarterly or annually (or weekly).

The presentation of content should be simple and clean, says Mathews. Live by Edward Tufte’s law of now more than six points per Power Point slide. Keep data tables as simple as possible; four columns by four rows are optimal.

Make the data easier to process by varying the shape of boxes, colour or even point size of the key numbers that should jump out.

Nudge

“Don’t just throw data at executives, give them actual recommendations,” Mathews urges. “Robots can pull numbers. You add value when you interpret and act on analytics.”

Presenting and assessing multiple options is good, just not too many or too few. A “lack of context” for information can often lead to misinterpretation of data and errors, Matthews says.

 

Iterate

Or as Mathews rephrases it, “keep throwing spaghetti at the wall.”

“It’s a conversation, not a lecture,” he says, so you need to keep repeating the key recommendations, with the renewed data to support it, on an ongoing basis. “Iterate, iterate, iterate.”

 

Originally written for and posted on In:fluencia Digital, a beta site created with Editions Infopresse to serve the Canadian online and interactive marketing, communications and media communities. The site’s development is in hiatus.

Posted by sutter or mckenzie at 9:16 AM EDT
Updated: Monday, 23 June 2008 4:15 PM EDT

View Latest Entries