May 07, 2002
When Linda Visintin's first email marketing test bombed (less than a 1% response rate to a free offer - ouch), she did not give up. Instead she kept on testing until she had a winning email marketing formula that works so well that other marketers would do well to steal ideas from it. Our favorite idea from this Case Study is that of putting the lead generation form in the actual email itself, so prospects do not have to click through to a landing page to respond. Apparently it is pretty easy too.
Four times a year, Avnet Hall-mark Division used to send out big direct mail campaigns to garner sales leads for the Oracle products that it is responsible for marketing as a value added reseller to small-medium businesses.
Avnet's in-house telemarketing team called each resulting lead to further qualify and sort leads, and then the sales team followed up to close sales.
The products sold for about $30,000 and have an average sales cycle of three-to-six months, so Avnet's direct mail campaigns (which are Oracle funded) were mailed right before the start of each business quarter to build up a pipeline of leads for that quarter.
This worked fine, until the boom era for tech sales ended a year ago.
Linda Visintin, Avnet's Marketing Project Leader says, "When the dot-com era was high, the direct mail response rate was hitting anywhere from one to three percent. With the era dropping off, we found that our direct mail response rate was declining as well. We needed to do something different in order to generate more response."
Like many B-to-B marketers, Visintin decided to test an email campaign.
At first, she did not want to invest too much in an untried tactic, so she ran a small test that she created in-house. "It was a straight text email that an end user would have to click on to view [the offer]. It flowed into a Flash presentation, and they had to do another click through in order to fill out a form."
Visintin did not have an in-house database of opt-in email addresses, so she rented a few email lists from a variety of sources including Yesmail and Harte Hanks.
The test bombed.
She says, "The response rate was less than one percent. We lost a lot of people through the click through." (Note: research shows you can lose a minimum of 10% of your visitors for each click you make them take before they accomplish a task.)
She also suspected that email success, just like direct mail, is largely dependent on the lists selected. "We didn’t know much about [email] lists at the time, so we learned a lot through doing the first campaign. We were trying to target b2b, and what Yesmail offered was more of a consumer list. The other thing we were looking for were responsive lists."
Visintin still believed email could general sales leads more effectively than direct mail if she could just get it right. Undaunted, she hired outside experts from Hotwire Interactive to create another test email campaign.
Step #1: Offer and creative
Historically, Avnet's direct mail campaigns had been successful with a free t-shirt giveaway, so Visintin definitely wanted to use that time-tested offer, but it needed something more.
Hotwire Interactive President Heather Clark says, "We were concerned that just using a t-shirt did not correlate closely enough with the product." She decided to offer both the free t-shirt, plus a free online information Kit about Oracle's product. (Link to sample kit below.)
Why would prospects be psyched to get an online "Kit" with information they could find themselves for free by just surfing Oracle's site? Clark explains, "Because Oracle's Web site is huge and there's tons of information. We distilled all that information down to a four-page site and made it very accessible and easy for people to find information they'd need to evaluate the product."
The Kit was specifically designed to enhance that feeling of an enclosed, easy-to-surf kit, rather than a full Web site. (Link to sample kit below.) Everything is enclosed within a rectangular box that fits nicely into an 800x600 resolution screen (the most commonly used screen resolution) without any scrolling.
Instead of using unsightly (and old-fashioned looking) frames to allow visitors to surf the Kit, Hotwire's design team decided to use Flash.
In the meantime, the copywriting team surfed Oracle's materials to pluck forth customer testimonials to be featured on each page of the Kit, that would reflect the benefits they were stressing in the rest of the copy, and be from peers of their specific target audience -- IT heads for medium-to-small companies.
Step #2: List selection
This time, the lists selected shared three qualities:
a. 100% guaranteed double-opt-in (in other words, each name had both opted in to be added to the list, and then confirmed their request by responding to an RSVP link in their email). Clark says, "There's been a lot of hype recently about privacy and making sure people are really comfortable receiving communications. And we knew with double opt-in there's a better chance that someone is going to respond to that and not be upset by it."
b. Highly targeted b2b names. The team chose to rent names from major trade publishers such as Cahners who could allow them to choose a very targeted demographic, including selections by company size.
c. Responsive names. Because the names came from trade publishers, and were double opt-in, they were more likely to be responsive names (people who had responded to emailed offers in the past) than names gathered from other sources.
Step #3: New test campaign
Visintin had diverted funds she usually put toward direct mail to this campaign. So, Avnet's next three - six month's sales were to a large part dependent on its success. Hotwire advised her to do something every smart email marketer should do: Run a test campaign before investing in a bigger one.
They decided to use 5,000 of the rented names to test two different factors:
Test A. Email subject lines. Aside from list selection, your choice of subject line is the most critical factor in an email campaign's success. (After all, nobody will see your wonderful offer and creative if they do not open the email in the first place.) The creative team tested three different subject lines:
1. Oracle9i - Just the facts (& a shirt)
2. How stable is your Internet infrastructure?
3. How to secure your infrastructure (& get a shirt)
As should be the case with all double opt-in list rentals, the email "from" line was dictated by and referred to the list owner, not Avnet. That is because the recipient opted in to receive mail FROM that list owner, and 'from' permission is not generally considered transferable (by prospects anyway).
Test B. Flash, vs. HTML, vs. text-only format. Since the prospects were all IT professionals at companies large enough to have T1 lines, but small enough to not be using pricey enterprise email software such as Lotus Notes which allows HTML email to come through but then renders it unreadable to the end-users, Clark felt free to recommend testing Flash and HTML versions.
However, she says, "Although we knew from a bandwidth perspective we weren't in danger of clogging up their pipelines, obviously all of the imagery was optimized, so it was definitely a readable size."
Our favorite part of the creative (see link to sample below) was that it included the response form in the email itself. Recipients did not have to click through to a form somewhere else to fill it out and get their free t-shirt!
Clark says, "We didn't even force them to click. They could sign up immediately in their email. It's not hard [to create] at all. If you can receive HTML email, you can receive a form in HTML. A lot of people don’t embed the form."
Why not? "I think it was just so commonplace to do a landing page, to do a click though to a place where you could fill the form out. A lot of people haven't switched gears yet."
It is worth noting you can not embed a form in text-email. In that case you must send people online to an HTML page for a form.
Step #4: Roll-out campaign
Responses came in so quickly that the team was able to see exactly which subject lines and format would be the best in only a few days.
Then they rolled the campaign out to all the lists they had rented. The rollout was staggered over a week or two in order so that Avnet's follow-up telemarketing team would not get too clogged with responses to deal with them in a timely manner.
Step #5: Telemarketing follow-up
After filling out the form, respondents clicked submit and then were taken to learn more at the online Kit. In the meantime, all responses were funneled to Avnet's in-house telemarketing team who quickly followed up within 48 hours.
Visintin explains the need for haste, "It gets cold after 48 hours, they suddenly don't recall what they've done."
People are hard to reach because of voice mail these days, so Visintin allowed up to four attempts per lead, and gave the telemarketers a careful script to use when leaving messages.
Based on prospects' answers, the telemarketing team classified leads into several different grades for the sales reps. Visintin says, "'A' is highly qualified, they have a timeline which could be zero to three months, and they have a budget. 'C' is the lowest quality and typically a literature request, maybe there's a project ahead but they don't have a timeline, they don't have a budget."
The rollout campaign pushed Avnet 143% over its lead generation goal for the last quarter of 2001. The leads received had a significantly higher ratio of hot 'A's to blah 'C's. Plus the campaign cost "significantly" less than a direct mail campaign to the same size list would have.
Visintin immediately began revising her plans for 2002, wherever possible moving budgeted funds from direct mail into email. (As a VAR, Avnet's marketing tactics are often constrained by rules set down by the supplier who pays for the campaigns.)
She adds, "Going forward, though, we're trying to educate them on why we need to continue email and how important it is. This campaign gave us a really good foundation on educating not only Oracle but also our other suppliers that we have, on using this competitive technology."
- Visintin was shocked when the Flash email creative was NOT the winner (HTML won by a landslide in the test). Clark says, "We found this was a very bad time to do Flash email. There were so many viruses going around small to midsize companies. So, the IT department doesn't allow people to see the plug-in in their email."
- Results from the initial test campaign revealed a big difference in open rates between different subject lines:
6% open rate: How to secure your infrastructure (& get a shirt)
3% open rate: How stable is your Internet infrastructure?
2% open rate: Oracle9i - Just the facts (& a shirt)
Why would a big brand name like Oracle not get the best response? Avnet's sales reps supplied anecdotal evidence to suggest that Oracle's fame for serving large enterprise-size companies worked against it in a campaign targeting smaller companies. Clark explains, "These companies saw the name 'Oracle' and they thought, 'Well, we just can't afford it.'"
- In part due to the addition of different lists in the campaign rollout, it actually pulled an even higher response rate than the test did:
9% overall open rate
4% overall response rate
Clark says she attributes much of the success of the high response rate to the fact that prospects could fill out the form in their email without clicking anywhere else.
- Troubles you wished you had: Despite high hopes, Visintin had not been expecting this great a response rate, so she had to scramble to get more telemarketers in place for that critical 48 hour follow-up call.
The team was able to reach about 50% of leads on the first attempt, and up to 70% of leads by the fourth attempt.
- Since the first campaign, the Kit has become a central giveaway for other Avnet programs because it furthers the sales cycle, and site stats show that prospects enjoy surfing it, especially the tech info sections.
In fact Oracle was smart enough to steal the Kit idea for their own site!
Avnet Hall-mark Division