by
David Kirkpatrick, Reporter
CHALLENGE
Hewlett-Packard faced the challenge many long-established B2B marketers have -- while its email list was large, it had many of those addresses in its database for a long time, and thus, engagement varied widely across the list.
In other words, many highly engaged recipients were getting drowned in a very large pool filled with long-time, but less engaged, subscribers. So HP decided to drain that pool in a targeted way to deliver more relevant content to the most valuable members of its audience.
The HP "Technology at Work" e-newsletter goes out to more than eight million subscribers across the world -- 50 countries and in 18 languages -- for both small-to-medium-sized businesses (SMB) and enterprise segments.
HP used two percent of its U.S. database to test dynamic email content based on user behavior and personas to increase engagement with its most devoted audience and this case study breaks that process down, step by step.
Cathy Howard, eMarketing Program Manager, Corporate Marketing, Hewlett-Packard, put it this way, "We are very careful not to spam subscribers, and very careful not to inundate them with emails that are just broad-based messaging. We want it to be an experience that is relevant and something that would be engaging to that subscriber."
CAMPAIGN
For the e-newsletter, HP has sign-up pages on its website, so the program is permission-based with some email recipients added to the list through sales reps’ recommendations for active prospects.
The portion of the email database that receives targeted dynamic content receives email that includes static elements everyone on the list receives along with special highlighted content that is based on that person’s behaviors and interaction with HP.
Step #1. Choose the segment for targeted dynamic content
"The way we came up with this behavioral pilot is because we are always trying to be smarter with our data, and there is a lot of information you can gather (from your email recipients), " Howard said. So HP indentified its most engaged email subscribers through a cookie tied to the email send and determined that two percent of the database should receive targeted email content based on their online behaviors.
Part of this list was also found through profile questions during the sign-up phase, such as company size of a new list enrollee. And HP uses an automated tagging process to track online interactions from the email send to determine the dynamic content each user receives.
Step #2. Determine the high-priority content
For segmenting, the marketing team broke the content out into three groups with two subgroups for each:
- Personal Systems Group
- Notebooks/laptops/tablets
- Desktops and workstations
- Imaging and Printing Group
- Multifunction printers
- Color, B&W laserjet, inkjet printers
- Enterprise business
"We have an editorial board that regularly drives the messaging to our segment audience," Howard explained.
She said each month the board would determine the most important message to be sent to each segment.
Howard added that when this effort was in its early stages, product promotion for the different business units was higher priority. After running this dynamic user-driven content email for some time, the team is now looking more at just providing information in the space.
"We are looking more and more to provide thought leadership content," stated Howard. "So, not just to be so promotion driven, because that is really not the purpose of (the Technology at Work newsletter.)"
There are static elements that are in every edition of the e-newsletter, and the highlighted content is dynamic and added to the email based on behavior and profile information.
Step #3. Continue to test and optimize
An important element to this marketing effort is continued testing and optimization.
Howard said the marketing team is testing things such as:
- Position of the behavioral content in the email
- Type of content -- product focused or thought leadership
- Time of day for the email send
- Subject lines
The marketing team also measures the subscribers’ activities on the behavioral and non-behavioral content.
"This is a process for us," said Howard, "We only have three months under our belt from this, so what we are trying to do is gather learnings and see what we can tweak to improve engagement."
She explained the process takes about two days. Marketing can pull a 48-hour report from the monthly newsletter to make changes for the next send.
Howard stated, "Most of your questions from an overall campaign perspective can be answered with that, and then we have a more of a deep dive analysis into the specific content to analyze which pieces the behavior group is more engaged with than the non-behavioral group. What they were clicking on."
She added, "It takes a little deeper dive into which piece of content is drawing (the behavioral group) in and how much time they are spending on each piece and if there are purchases related to it."
Step #4. Determine the internal process
The marketing team has a fair amount of autonomy with this effort to do things like:
- Sample testing within the database
- Choose images related to the email sends
- Write titles in the email
- Determine where content is going appear
At the same time, the team has to answer to stakeholders in the company who can override those decisions, and Marketing has regular editorial meetings with the internal stakeholders to make sure they are aware of the choices the team is making and that there is a consensus on those choices.
RESULTS
Howard said the marketing team tracks all the typical email campaign metrics with this effort, such as volume sent, open rates and unsubscribes, along with continually testing subject lines.
The campaign review is more of a content performance review with the different business units because each is interested to learn how their content performed and use those learnings to produce the next content piece.
The metrics on the limited part of the database that receives the dynamic content is impressive:
- 300% higher open rate
- 600% higher clickthrough rate
- Generates as high as 18% of overall traffic
Howard said, "I think one of the things I am striving for is improving how we qualify this group. I think the group is larger than what we are qualifying, and we are a little bit limited on how we are able to gather that knowledge. I think that is improving and will improve going forward."
She added that the dynamic content sent to the targeted audience has moved from a focus on product promotion to more thought leadership and educational content.
Cathy Howard, eMarketing Program Manager, Corporate Marketing, Hewlett-Packard, will present this case study at the upcoming MarketingSherpa B2B Summit 2012, to be held August 27-30, in Orlando.Creative Sample
Example of two versions of "Technology at Work" e-newsletterSources
Hewlett-PackardPepper – HP’s email marketing agency
Related Resources
Combining Offline and Online Marketing: How HP built relationships, boosted sales 2,050%How Hewlett Packard Uses Email, Online Surveys and Web Sites to Grow Customer RelationsUse Dynamic Content and Testing to Optimize Newsletters: 4 Strategies to Create Customer DialogsEmail Marketing: Dynamic images get 5 opens per reader on a single emailOne-to-One Marketing: The true promise of Dynamic Offer-Content Customization
