It’s vital to get the first couple of words exactly right when
writing effective web headings and links.

The first two words have a huge impact on whether or not people
will click on a link according to a new study by Jakob Nielsen.
This basically confirms the findings of a 2004 Eyetrack study
from the Poynter Institute that found that, “Most people just
look at the first couple of words-and only read on if they are
engaged by those words. For headings-especially longer ones-it
would appear that the first couple of words need to be real
attention-grabbers if you want to capture eyes.”

The Jakob Nielsen study tested 80 people and found that they
typically see the first two words in a link. The study tested
links from websites such as AT&T, Intel, Dell and UK

The best links in the study:
* Used plain language
* Were specific and clear
* Used common words
* Started with the essence of the message
* Were action-oriented

The worst links in the study:
* Used bland, generic words
* Used made-up words or terms
* Started with after-dinner-speech-

introduction language

There is nothing worse on the Web than welcoming people, and
telling them about how you’re so delighted to announce the
launch of, or about how on your website they will be able to
find, or about how it’s now even easier, or about how you’re
introducing, launching, or already in an orbit of hot air.

Web content is brutal and to-the-point. Lead with the need.
Don’t get to the point. Start with the point. Remember, your
website is the context, so you should never start a link,
heading or sentence with your organization’s name. The website
visitor knows your name. They’re on your website, after all.

Links and headings are very similar in their function. They act
as signposts, as promises. Do not mislead or over-promise. One
of the most common mistakes in link writing is that the link
does not deliver on the promise. The links says “buy stamps
here” or “download now”, and four clicks later you’re still not
able to download. Now means now, not five clicks, two forms and
four minutes later.

Links and headings should be no more than 8 words. We did a
study recently of 500 web news headings. Over 70 percent of them
were 8 words or less. Avoid putting links in sentences because
this makes it harder to read the sentence, and harder to
understand what the purpose of the link is. Instead put links on
separate lines after the content.

Avoid PR, advertising and marketing gibberish at all costs.
Please, please never, ever write anything like “solving
tomorrow’s problems today” or “excite your senses” or “the human
element is providing comfort every day.”

Don’t be smart, clever, obscure, vague. Be clear, compelling,
concise, and always focus on what your customers really care
about. And remember, what your customers really care about is
very often not what you really care about.

First 2 Words: A Signal for the Scanning Eye

Gerry McGovern



Most web content is overwritten; too much content, too much
context, not nearly enough focus on the action. Unfortunately,
we’re taught to write this way.

How often are you presented with content on the Web that begins
something like this: “Exciting, compelling, and effective user
experiences result in high levels of customer loyalty,
satisfaction, and referral.” On the surface, this seems like an
okay sentence. It’s how we’re taught to write: set the scene,
establish the context.

However, it’s utterly useless. It’s like saying: “Every business
is an end-to-end network of interrelated people and processes.
The more seamless and flexible the network, the more successful
the business.” Or: “Your people are your most valuable resource.
They contribute to the success of your company.” Or: “Even
during the best of times, companies are always looking for ways
to trim costs, optimize processes, drive efficiencies, and
create greater value for their clients.”

The problem with the above sentences, other than the fact that
they are utterly useless, is that they are utterly useless. (Not
to mention the fact that they are utterly useless.) They don’t
tell you anything you don’t already know, and they give you no
real sense of what the product or service is actually about.

If someone is at your website they already have the context.
They have made a deliberate decision. They are in an active,
doing mode. They want to dig deeper, compare, price, to get
detail, detail, detail.

Write web content from an elevator pitch perspective. Your
customer has walked into the elevator, the doors have closed,
they turn to you and say: “Convince me before the next stop to
buy your product.” Design your website from the ‘I badly need to
go to the toilet’ perspective. Your customer needs to act and
act quickly. That’s the Web.

You’re proud of your website but pride comes before the click of
the Back button. Anything on your website that puffs your ego,
that makes you smile, that you think is really cool-remove
immediately. The content that you’re in love with-and so proud
of-is nearly always the content that drives your customers away.

There is far too much content written for the English teacher or
the English exam you crammed for. You want to impress. You want
to show off all the clever things you know. You want a
beginning, middle and end. You want to tell them what you’re
going to tell them, tell them, then tell them what you’ve told

A normal person sees a link called “Where’s my refund?” and
thinks that if they click on this link they’ll be able to answer
that question quickly. But a classically trained English student
who wrote the link thinks that when the person clicks on the
link they should be given this sentence. “You filed your tax
return and you’re expecting a refund. You have just one question
and you want the answer now: Where’s my refund?”


To Nonprofits Seeking Cash, Facebook App Isn’t So Green

Washington Post Staff Writers
Wednesday, April 22, 2009

It seems foolproof: nonprofits using the power of the Internet to raise money through a clever Facebook application. After all, the Web earned gobs of cash for Barack Obama’s presidential campaign. And besides, going online means sending fewer fundraising letters, which makes it appealing to penny-pinchers and environmentalists alike.

But it turns out that approach doesn’t always work. The Facebook application Causes, hugely popular among nonprofit organizations seeking to raise money online, has been largely ineffective in its first two years, trailing direct mail, fundraising events and other more traditional methods of soliciting contributions.

Only a tiny fraction of the 179,000 nonprofits that have turned to Causes as an inexpensive and green way to seek donations have brought in even $1,000, according to data available on the Causes developers’ site. The application allows Facebook users to list themselves as supporters of a cause on their profile pages. But fewer than 1 percent of those who have joined a cause have actually donated money through that application.

The data conflict with the lessons many nonprofits took from Obama’s presidential campaign: that a well-run organization could raise huge amounts of money online. The problem is, nonprofit fundraising requires considerably more outreach than many political campaigns, which do not require as much relationship-building because they revolve around highly visible candidates.

Research shows that Internet and e-mail are generally considered the least successful nonprofit fundraising techniques, according to a report by the Center on Philanthropy at Indiana University.

“The prevalent fantasy among nonprofits during the early days of the Web was that a random person would come to your Web site, see that they could donate, and donate a million dollars,” said Aaron Hurst, chief executive of a California nonprofit, who has blogged about the ineffectiveness of Causes. “But that wasn’t true then and it isn’t true on social networks.”

Since it was launched in 2007, Causes on Facebook has become the leader among a growing number of social networks — including Twitter, MySpace and Gather — used by nonprofits, which have been forced to find new ways of developing resources as contributions from wealthy donors and foundations decline during the recession. Causes is free for nonprofits but it costs them staff time to develop and maintain.

Data available from the Causes developers on Facebook show the application’s meteoric rise since its founding. More than 25 million of Facebook’s 200 million worldwide members have signed on as supporters of at least one cause, making it the third-most popular of the more than 52,000 applications on the site.

But just 185,000 members have ever contributed through the site, which sends credit card transactions on Facebook to the Bethesda-based Network for Good to distribute. The median gift through Causes is $25. The majority of Causes’ participants have received no donations through the site.

The median charitable donation through more traditional means is $50, according to the Center on Philanthropy.

The idea behind Causes was to take advantage of the vast circles of online friends connected through social networks to reach potential donors and volunteers on a more personal level. People will donate money, albeit in small amounts, to help a cause that a close friend or colleague supports, the application’s developers say.

“People are much more altruistic if they get social credit for it,” said Joe Green, one of the founders of Berkeley, Calif.-based Causes, who said the application has raised $7 million overall. “The social incentive is to show on your profile how many volunteers you’ve recruited or how much money you’ve raised.”

Green acknowledged that the amount of money raised per member is not large. But he said that Causes allows small groups to raise money they otherwise wouldn’t.

He said Causes raises almost $40,000 a day across its groups, up from $3,000 a day a year ago. “The biggest successes have been tiny nonprofits who don’t have the name recognition of the big guys.”

But in the majority of cases, that theory hasn’t translated into significant dollars. Fewer than 50 of the 179,000 groups on Causes have raised $10,000, and just two — the Nature Conservancy and Students for a Free Tibet — have cracked the $100,000 mark.

Hurst jumped on the Causes bandwagon shortly after it launched. The chief executive of the San Francisco-based Taproot Foundation, which recruits professionals to perform pro bono work for nonprofits, he figured creating a Causes page would bring in plenty of new donations. Taproot spent about $3,000 in staff time developing the page, Hurst estimates.

Six months later, Taproot’s Causes page had netted $30, all from existing donors. The group’s page still exists, but the staff has largely abandoned its upkeep.

In 2007, Eric Ding, a Harvard postdoctoral researcher, used Causes to create the O Campaign for Cancer Prevention, which raises money for cancer research at a Boston area hospital. The group now has 4.6 million members and has raised nearly $85,000, less than 2 cents per member.

Nevertheless, Ding, 26, was very satisfied with the effort. “That’s a lot of money that didn’t exist before,” he said, “especially with recent funding cuts for cancer research.”

Alan J. Abramson, an expert on philanthropy at George Mason University, estimates that less than 3 percent of all fundraising is done online. “Nonprofits raising money through the Web is growing, but it’s still pretty small,” he said.

Even e-mail campaigns are generally more likely to raise significant amounts of money than Causes, according to a comparison of Facebook’s data with widely accepted philanthropic benchmarks. Those data show that 1 percent to 3 percent of a nonprofit group’s e-mail list would donate money when solicited, at an average of about $80 per person. That would have brought Ding’s cancer research group $3.7 million, more than 44 times what he made on Facebook.

For the Nature Conservancy, which is the top fundraiser on Causes with $198,000, social networking was never primarily about raising money. The group has four staff members devoted to representing it on social networks, but uses Causes mostly to circulate news stories and event announcements.

“I definitely think it’s first and foremost a tool for brand and reputation,” said Sue Citro, the group’s digital membership director. “It definitely does more for influence than for fundraising.”

For Washington resident Elliott Bisnow, who concluded a weeklong fundraising campaign for Nothing but Nets on Twitter and Facebook two weeks ago, social networking was an experiment in recruiting friends to a cause that he’s passionate about: preventing malaria by distributing mosquito nets to African children.

About 280 people donated a total of $8,000, far less than Nothing but Nets collected at a recent fundraiser, but more than Bisnow could have donated himself.

“It’s an experiment that shows social media is a great way to touch more people,” he said, “but it may have a ways to go in terms of raising money.”


European Internet Stats

Current European growth trends:
* 9 hours per week spent on the web in 2008, up 27% from 2004 – more time than
people spent reading print media, watching movies or playing video games

* North South divide in Europe: Nordic countries have an internet penetration rate of 76% on average, compared to 45% in Southern Europe

* Internet consumption set to overtake traditional TV in June 2010 – 2.5 days per month spent on internet versus 2 days on traditional TV: but does not signify decline in TV- simply a shift in how it is consumed

* Online video established as most popular online audiovisual entertainment application, with more than 1 in 4 (28%) Europeans and over 300 million people worldwide watching short or full-length videos on the internet

* Time spent on the internet using a PC will drop from 95% today to 50% over the next 5 years


Web Content Management

The first time I heard the phrase “content management” I made
the pretty silly assumption that content management was a
discipline focused on how to professionally manage content. How
wrong I was. Content management was, and to a large extent still
is, about content management software.

The school of content management brought us such developments as
portals, customization, personalization, and distributed
publishing. These management-free, technology-driven solutions
have led to public websites and intranets teeming with poor
quality, badly organized, out-of-date content.

What do you get when you personalize crap content? Personalized
crap content. What do you get when you distribute publishing
rights to people who can’t write, don’t care about what they
write, think metadata is a country bordering Outer Mongolia, and
will never, ever review or remove what they publish? You get the
website you deserve.

Why do so many organizations think they can solve the problem of
customers finding stuff on their websites by simply buying a new
search engine? No extra staff. No management of the search
process. Magic mushrooms may exist but a magic search engine
certainly doesn’t.

Technology is important, even critical, but we still need
quality people to manage websites if we want those websites to
deliver value to the organization. Web managers’ number one task
is to develop a deep understanding of customer web behavior.
There is no better way to do this than to observe customers as
they seek to complete top tasks on your website.

Recently, Amazon got in a lot of trouble when a ‘technical
glitch’ or ‘cataloging error’ resulted in thousands of books,
including many gay and lesbian books, becoming much more
difficult to find on its website. “This whole mishagoss could
have been easily avoided but for the one thing Amazon (not to
mention a whole lot of other online entities) is notoriously
remiss – actual human-on-human customer service,” MSNBC’s Helen
Popkin writes.

“Like many of its online compatriots, Amazon wants to automate
everything,” Ann All writes for ITBusinessEdge. Amazon is a
truly customer-centric organization that continuously invests in
understanding its customers, but sometimes even they get it

To understand web self-service it is much more important to
develop an understanding of human behavior in an online
environment than to master any technology.

Gerry McGovern


Content Quality Checklists

In my experience, a common misperception of the evaluation of content quality is that its scope is limited to the correction of typos and grammatical errors. Correcting spelling and grammar only scratches the surface. To truly consider content quality, we need to examine its quality along several dimensions. Consequently, the content quality checklists that follow cover everything from usefulness to voice to accuracy.

  • Usefulness & Relevance:
    • Does the content meet user needs, goals, and interests?
    • Does the content meet business goals?
    • For how long will the content be useful? When should it expire? Has its usefulness already expired?
    • Is the content timely and relevant?
  • Clarity & Accuracy:
    • Is the content understandable to customers?
    • Is the content organized logically & coherently?
    • Is the content correct?
    • Does the content contain factual errors, typos, or grammatical errors?
    • Do images, video, and audio meet technical standards, so they are clear?
  • Influence & Engagement:
    • Does the content use the most appropriate techniques to influence or engage customers?
    • Does the content execute those techniques effectively?
    • Does the content use too many or too few techniques for the context?
  • Completeness:
    • Does the content include all of the information customers need or might want about a topic?
    • Does the content include too much or too little information about a topic for the context?
  • Voice & Style:
    • Does the content consistently reflect the editorial or brand voice?
    • Does its tone adjust appropriately to the context—for example, sales versus customer service?
    • Does the content convey the appropriate editorial and brand qualities?
    • Does the content seem to have a style? If so, does the content adhere to it consistently?
    • Does the content read, look, or sound as though it’s professionally crafted?
  • Usability & Findability:
    • Is the content easy to scan or read?
    • Is the content in a usable format, including headings, bulleted lists, tables, white space, or similar techniques, as appropriate to the content?
    • Does the content have the appropriate metadata?
    • Does the content follow search engine optimization (SEO) guidelines—such as using keywords—without sacrificing quality in other areas?
    • Can customers find the content when searching using relevant keywords?

A Few Caveats

While I think these content quality checklists are a good place to start, it’s important to note several caveats relating to their scope and appropriate use.

  • As for any heuristic evaluation or competitive analysis, the expertise of the person evaluating the content is as important as the heuristics he or she uses.
  • An expert opinion is still an opinion. It is also important to consider other indicators of content quality such as Web metrics, usability testing, and customer feedback.
  • These checklists do not entirely fit user-generated content or user assistance content. However, I think variations of these basic checklists might be useful for those contexts.
  • These checklists certainly do not replace other important content strategy tools such as detailed content inventory and analysis.

Why Content Quality Matters

“If you already care about content, you know instinctually that its quality matters.”

If you already care about content, you know instinctually that its quality matters. Personally, when I encounter bad content, my blood pressure starts to rise, and I feel an uncontrollable urge to fix it. However, not everyone instinctually cares about content quality. (If they did, bad content would not be so rampant.)

What if you have to convince people who are oblivious to content quality that they need to improve the quality of their content? Persuading them will require more than a heuristic evaluation or a competitive analysis. Fortunately, some experts have already made compelling cases for content quality, so you don’t have to start from scratch. You can build on their convincing arguments:

  • Content is a strategic brand asset. If the purveyors of the brand or marketing are your audience, this argument will resonate. Content is a significant part of the brand experience, and we should treat it as such. Improving content adds luster to the brand. Kristina Halvorson and Joe Pulizzi make this case effectively. [3]
  • Content is a major part of the user, or customer, experience. If those who are responsible for user experience or customer service are your audience, this argument will garner attention. Redish’s Letting Go of the Words can help you make this argument. Halvorson also makes a clear case that content is a user experience issue. [4] It’s not just some arbitrary substance that fills our designs.

Ensuring Content Quality: Beyond the Style Guide

“Engagement is a metric that suits content well.”

Let’s say you’ve made a winning case for content quality and successfully improved its quality on a project. Inevitably, the question of maintaining the quality of content will next arise. I’ll now highlight a few ways of effectively maintaining content quality. This topic is rich and merits more discussion, but this list will at least help you move forward:

  • testing content with users—Incorporate questions about content in your user interviews, focus groups, usability tests, and surveys. We don’t discuss testing of content enough, and I plan to discuss it more in upcoming columns.
  • monitoring content metrics—Engagement is a metric that suits content well. For content that’s meant to support conversions, tracking whether conversions increase after improving content is important. For example, InterContinental Hotels Group conducted a pilot comparing their existing hotel content with improved content, which was professionally written and photographed. Everything else about the user experience was identical. With the improved hotel content, the increase in conversions was so significant it resulted in an immediate decision to improve all hotel content. [5]
  • establishing governance—Have a group of stakeholders from across the company or organization meet regularly to oversee major content decisions.
  • applying the publishing model to content—The publishing industry, for the most part, knows how to develop quality content—despite occasional disappointing efforts. (See Joe the Plumber as author.) Jeffrey MacIntyre articulates the publishing model well. [6] For major content efforts, a publishing structure and related tools—such as an editorial calendar—are a natural fit.
  • incorporating content guides, standards, and tips into CMS workflows—I would love to see guidelines for content quality incorporated into CMS (Content Management System) workflows. Including some contextual Help or quality checklists would provide effective reminders, preventing the people who approve content from forgetting what the content guidelines are. Publishing guidelines in a hard-to-find document probably won’t help much in the day-to-day grind of content publishing.
  • maintaining the metadata—I am not a metadata expert, but I know that, as the semantic Web progresses from dream to reality, metadata is becoming even more important to content quality. Rachel Lovinger demystifies metadata, noting important strategies and tools. [7]
  • hiring employees, consultants, and agencies who care about content—Whether you need a lone content evangelist, an outside agency, or your own content team, hiring people who have an instinct for and expertise in content will go a long way toward preserving its quality.


[1] Redish, Janice (Ginny). Letting Go of the Words: Writing Web Content That Works. St. Louis, MO: Morgan Kaufmann, 2007.

[2] Jones, Colleen. “Content Quality: It’s More Than Fixing Typos.” Content Strategy Consortium at IA Summit 2009, March 18, 2009. SlideShare. Retrieved March 26, 2009.

[3] Halvorson, Kristina. “Content Strategy: The Care and Feeding of Your Most Important Brand Asset.” Conversations About the Future of Advertising, March 9, 2009. Retrieved March 26, 2009.

[4] Halvorson, Kristina. “Content Strategy: The Mania, The Myth, The Mayhem.” SlideShare,February 2008. Retrieved March 26, 2009.

[5] Jones, Colleen. “Usable, INFLUENTIAL Content: We Can Have It All.” IA Summit 2009, March 21, 2009. SlideShare. Retrieved March 26, 2009.

[6] MacIntyre, Jeffrey. “Publishers and Content Strategy.” Content Strategy Consortium at IA Summit 2009, March 18, 2009. SlideShare. Retrieved March 26, 2009.

[7] Lovinger, Rachel. “Metadata Strategies and Tools.” Content Strategy Consortium at IA Summit 2009, March 18, 2009. SlideShare. Retrieved March 26, 2009.


Badci Journalism skills today

Today I got an e-mail from a journalism undergraduate with a few basic-sounding questions that I could answer quickly. But when I looked at my answers, I realize they have some more profound implications then she was probably expecting:

1. What is the most important skill you use in your posts on the Web?

Having a good sense of what’s likely to be interesting to the people I’ve connected with (or who I’d like to connect with), and why.

2. In your opinion, what is the most effective way to tell a story online (pictures, text, sound, video, etc.)?

You should know how to use all these tools and know the people/communities you want to connect with, and what their media preferences are (both for media content type, and the tools they tend to use most). Then tell your story in a form that will work best for them.

Stories don’t exist for their own sake, and you are not your audience. It only works if you really connect with people, and that means taking them into account from the start.

3. What is the hardest part about being an online professional?

Anyone these days who’s doing any kind of media work is inherently an online professional in some way, directly or indirectly. People who deny that or try to avoid it make their own careers impossible.

4. What core skills do you think every journalism major should have?

Many, but the most basic one is: How to define and connect with communities. This is the basis of all media activity, including journalism — but too often it’s taken for granted and not studied and understood in its own right.


Five “Don’ts” of Nonprofit Website Design

Your nonprofit’s website can be a powerful, strategic tool. Unlike the days of brochure-like static pages, home pages now have the potential to win over potential supporters and reaffirm the folks who already know you.

When you boil it down, websites do not just sit on a server-they are action-oriented. They persuade and (hopefully) convert. And for the latter-in terms of raising money online-your website has the most potential with two groups of donors: new donors and impulse givers.

When these newbies visit your website, what do they see? What’s their experience with your navigation and donation processes. According to recent research, nonprofits could be leaving as much as 10 percent of their online revenue on the table simply due to two website usability issues: content and design.

Read on for the five content and design flubs to avoid when you aim to convert browsers into donors:

  1. A lack of call-to-action. The number one thing to avoid when asking for donations on your website is to forget to make the ask! (Yes, this is also on our list of “website do’s,” but it’s important enough to mention at least twice.) If you don’t ask for donations, website visitors might think you don’t need them. Yes, it’s almost laughable to us in the nonprofit world, but Web-savvy surfers assume that if something’s missing, it’s intentional.
  2. Jargon breath. (No, this has nothing to do with the take-out you had for dinner last night.) “Jargon breath” refers to a tendency by communicators-particularly in the nonprofit sector-who rely on a particular vernacular of terms to try to educate others about their mission and programs (“services,” “accessible,” “at risk,” etc.). But, as Tom Ahern, an authority on effective donor communications,  so lovingly points out, “jargon just conjures confusion and blank mental screens.” Go over your website copy with a fine-toothed comb, and perhaps a friend who doesn’t work at your nonprofit, and flesh out what your organization really does. Show people and explain in real terms. (Learn more below “Related Articles” when you scroll down.)
  3. Unintuitive navigation. This fix could be as simple as changing the text on your navigation’s buttons. Is your donate text hidden behind an “about us” button? Are you asking people to “join” you, but your home page doesn’t indicate that you’re a membership organization? It may be time to break out the “Grandma-intern-or-significant-other” test: Sit someone down in front of your website and watch them navigate around your site. Try to quiz them to find certain areas and probe them for feedback. Whether the person’s a Web designer or a teen who spends an hour on Facebook every day, you’ll be sure to glean some important info.
  4. Inconsistency with the mother ship. Of the 1.5 million nonprofits in the U.S., most are small, one-location shops with small budgets (but big hearts). However, if your nonprofit is part of a national or international network, you’ll want to avoid completely flying off the brand handle when you’re working on your local website. You may have great design resources and a quirky new take on your organization, but you want to make sure site visitors (i.e. potential supporters) easily make the connection between your site and the national one they may be familiar with already.
  5. Confusing, third-party donation processing. Make it as easy as possible for supporters to donate to your cause. They’re trusting your organization with their hard-earned cash, as well as your website with their credit card. According to Network for Good’s own research, branded donation pages like Custom DonateNow bring in a higher average donations ($125) and improve the donor experience. (Improvement refers to shortening time to complete the transaction, lessening the number of clicks and giving the feeling that a supporters has not left the nonprofit’s website.)

How much email is too much

That’s the 25 million dollar question, especially for an advocacy group like Amnesty International USA (AIUSA). We’re the nonprofit group that protects people wherever justice, freedom, truth and dignity are denied. Talk about a broad mission.

It’s no surprise then that when I first joined AIUSA as the Managing Director of Internet Communications, we were sending out 2, 3 or more emails PER DAY. Yes, we segmented, but trying to build suppressions, queries and code emails to send 2 to 3 emails per day was a nightmare and not always effective. The online team at that time was primarily seen as a group of glorified tech-monkeys who would take copy and email it to our list. The quality varied from downright embarrassing to just OK, but still really wonky and dry.

I knew immediately a couple things needed to change: 1) the online team needed to be key decision makers on email; 2) our volume had to decrease; 3) the quality of the writing had to improve.

Because we had the keys to the tool that actually sent the messages, I began acting like we had the authority to do things differently. The first thing I did was rewrite email copy sent to the online team, and I asked the other online staff to do the same. Programs didn’t like us rewriting their copy, but I was persistent, and told them that we knew how best to write emails meant to mobilize online supporters. Our writing at the time primarily focused on having great hooks that were timely, and focusing on individual stories that could humanize our issues. It probably took a year before other departments got comfortable with our expanded role.

To address our email volume, I first measured how many emails our average subscribe received and compared it to other advocacy groups. We were at the very high end, sending most subscribers between 19 to 25 emails a month. Yikes!

I used this comparison, along with some research from M+R that showed reduced email volume improved response rates. Admittedly, the research wasn’t so cut and dry, but it was enough to make a case.

Then I put together a set of email guidelines that gave allotments out to the staff in charge of: fundraising (usually 2x a month), priority campaigns (up to 8x month), and non-priority programs (up to 4x a month). There were a few other emails that could get on the calendar (event invites, registrations) too.

This approach forced the individual programs and campaigns teams to go lobby their supervisor, not the online team. I remember when we proposed the new structure for email communications, there were all sorts of predictions about how we’d no longer be able to do our work, that our campaigns would fail, and the world would probably end.

A year into it, we found that most of the objections were exaggerated. However, there were some important emails that these guidelines didn’t allow, like sending super targeted actions to key targets during key moments, or thanking people after we achieved something. So we adjusted and loosened the guidelines to allows for these important types of emails.


Our first set of guidelines were probably more like a sledgehammer than a scalpel, but they were critical to changing the organization’s inaccurate view that high volume, low quality was an OK way to use this scarce resource. We’re now about to release our third iteration guidelines and these are much more strategic.

Ben Brandzel, formerly with MoveOn, Avaaz and the Edwards campaign, conducted a 5 hour training with us on what makes a great email. The gist is that email really is only effective when you can clearly articulate a crisis, an opportunity (crisitunity), and a theory of change (how taking action now will resolve the crisitunity).

Some examples of crisitunity and theory of change:

  • Good crisitunity: Monks are being killed in Burma <crises> and China has the power to stop it. <opportunity>
  • Bad crisitunity: Violence against women threatens the fabric of society.
  • Good theory of change: China is Burma’s only real ally, and if they pressure the junta, Than Shwe will have to back down. It’s up to us to call on China and make sure that they do. So we’re launching a petition today and broadcasting your signatures through an ad in the Financial Times – with a huge circulation among the power brokers of Bejing.

Bad examples of theory of change:

  • Missing: “Global poverty is terrible, and we’ve launched a petition to stop it.
  • Impossible: “George Bush has staked his presidency on privatizing social security. So we’ve launched a petition asking him to stop.”
  • Obscure: “Climate change threatens us all, and we’re working night and day to stop it. Please contribute to keep our campaign going.”

Based on this model, I am now proposing that 80% of all our email be reactive, and 20% proactive. I’m not setting specific allotments but telling campaigns and programs that if they can show me a great crisitunity and theory of change, we’ll send it to the full list.

Along with this reactive email, the programs will be able to choose about one moment a year when they can proactively push a major project via email, and we’ll send out alerts to the full list.

Anyone who responds during these moments, or during full-list reactive actions, can be considered part of that issue’s segment. This segmented list can be occasionally accessed during other non-reactive times when they really need support.

The biggest lessons we’ve learned on this journey is that emails that are highly opportunistic, that can clearly show the importance of the moment, in very specific terms, as well as a clear advocacy strategy, perform leagues ahead of other emails. My feeling is that every email needs to meet this bar, otherwise, email isn’t the right tactic to achieve the stated goal.

*This article was written by Steve Daigneault who is the Managing Director of Internet Communications for Amnesty International USA.