PURLs, GURLs or CURLS – Personalization versus Relevance

By | June 4, 2009

Personalized URLs in combination with direct mail or email campaigns have emerged as a valuable marketing tool. A Personalized URL (PURL) is a web page or microsite that is tailored to an individual visitor. A common example of a PURL is a web page with a URL such as www.retailstore.com/johnsmith, where “John Smith” is a target prospect or customer who receives a direct-mail or email communications that encourages him to visit the web page. For each recipient within the campaign, the web address is unique and personalized to the recipient. The content of the web page (PURL) should also be tailored to each recipient through the use of variable fields allowing pages to be linked to a database that contains information about each potential visitor.

PURLs are a stark contrast to a GURL, or generic URL, which is more common in direct mail and email campaigns. For example www.phonecompany.com/SummerSavings where PhoneCompany is offering the same call to action to everyone who receives a campaign offering discounts to new customers during a certain time period (I think I’ve received 3 this week!) If the offer is compelling and the design is effective, a GURL can work pretty well. But a variety of research (PODi.org, Aberdeen, Jupiter) tells us that personalization offers a significant lift over generic campaigns, but personalization comes at a price.

While software can be purchased to allow marketers to generate PURLs at lower cost per URL, more commonly, marketers are using 3rd party services to generate the PURLs and track response. For third party services, costs range from .05 to .50 per PURL per campaign depending on volume.

Before spending the money, consider that the degree of personalization that can be created on PURLs is highly dependent on the amount of data your organization has gathered about its target audience. When good data is available, keep in mind that an overly detailed and personal level of information in a PURL may lead a visitor to feel his or her privacy is being invaded.

Here are the biggest drawbacks I see with PURL implementations -beyond the “Creep Effect” above:

  1. Many marketers are so enamored with the idea of putting John Smith’s name on a web page that they don’t spend the time making sure that the page meets John Smith’s needs;
  2. Often, John Smith’s needs aren’t radically different than, say, Karen Jones’ needs. In fact, there may be only a handful of drivers that make the campaign relevant to the individual.
  3. Relevant messages are more effective than those that are only personalized – so a PURL without a relevant message is a poor use of resources.

The real driver of response rates is relevance rather than personalization. PODI.org cites research indicating that relevant campaigns generate, on average, 300% better response than those that are simply personalized. The level of customization to make something relevant to the recipient is usually not 1-to1 or even 1-to100. It may be more like 1-to-10,000.

Here is where Customized URLs or what I call “CURLs” come in. You can tailor a particular web page to the level necessary to make it highly relevant to each group of recipients. Perhaps tailored by length of relationship, level of purchase power, industry or region. You can still personalize the direct mail or email campaign that directs the recipient to the website. For example,

“Dear John,

Pharmaceutical executives like you spend an average of $4,000 per year on rental car expenses. Visit www.rentalcarco.com/pharmaexecs, for discount programs and other tips that can help you cut your costs by up to 20%.”

Assuming that the message was being sent to a list of 100,000 prospects and tailored for 10 different industries, this campaign could be launched with out any specialized PURL generating software or services. In contrast, the same campaign with PURLs would generate 100,000 individual microsites at a cost of $10,000 to $20,000 before the cost of the direct mail or email blast.

So save some money and still get results:

  • CURLs are at least as effective as the typicall PURL implementation at a fraction of the cost
  • Spend your time and budget coming up with great, relevant offers and supporting creative to increase response
  • If you generate the CURLs internally, make sure you have a method for measuring response

Whether you use PURLs, CURLs or GURLs using combining direct mail with online media to get your message across and measure response is a big step in the right direction.

This article originally appeared at Insight Forums. Insight Forums is an analyst and consulting firm that establishes business communities for communications professionals in highly-regulated industries.

Share this post


10 thoughts on “PURLs, GURLs or CURLS – Personalization versus Relevance

  1. Ryan Lou

    Whether it is Purls, Gurls or Curls, it is also important to start thinking of campaigns not just as a single email to purl or Direct mail to purl campaign. Rather, it should be a continous 2-way conversation between a brand and its prospects.

    Personalization and relevance that these channels affords, allow marketers to continue to keep the conversation real. For example, personalization doesn’t just mean using the first/ last name or creepy personal details.

    It could be something like “John, in the last message you said that access to credit was a problem for you, here’s a payment scheme that we know will work for you”.

    When campaigns are also continuous, then personalization and relevance will work like a charm and not be creepy. You will be remembering what the prospect said rather than using more and more creepy personal data.

    Are continuous campaigns more expensive? Probably, but remember, you will be tracking response down to the individual. This means you can ensure you spend 80% of your marketing budget on the 20% of prospects that will likely end up as customers.

  2. MichaelJ

    Very good post.

    My bet is that the research would show that the “Creep factor” as in “how did you get my name” is a bug, not a feature. After years of personalization in direct mail and the concerns about privacy, I have to believe that the micro reaction to seeing one’s name or any personal information on a web page is “What else do they know about me?”

    On the other hand, coming to a page that has a selection of “interesting to me” can be a delight. It creates the “How did they know that “people like me” would be interested in all this cool stuff.

    On the other issue of the marketing campaign. In the days before the possiblity of real time metrics of engagement, the only indicator of success was the sale itself. The paradox is that sale is the end of the process, not the beginning of the process. It is better thought of as the conversion rate.

    Most printing salespeople understand that getting to conversion can be a long process depending on the nature of the value proposition. Retail stores have always understood the importance of traffic and have well defined measures to relate traffic to conversions.

    The new reality is that spam – either in the form of email or paper – has lost it’s power to be the surrogate for “traffic.” But what I’ve been calling Clickable Postcards and Newsletters might be newly appropriate ways to start and nourish the conversations that lead to lifetime customers and continuing conversions going forward.

  3. Elizabeth

    Nice additions Ryan and Lou. Thanks.
    On the “Creep Factor” – just adding a name isn’t creepy (although sometimes it doesn’t add much value.) but organizations need to put thought into creating the dialogue (or “continuous campaign” as Ryan puts it) so that personal data is used in a natural, conversational way that does not shout out “hey – look at all the stuff we know about you!” It probably seems obvious but I’d rather err on the side of stressing how important it is to use personal data respectfully.

  4. Colin

    Great article

    It’s all about ROI
    It can now be shown that adding personalization and relevence will yeild the best results. Clearly if done right the PURL option will yeild a better result. The key is all about the data, continuing the conversation and gather more data along the way. The ability to react to specific questions and gather intel is key.

  5. Joe Barber

    Great post. Couldn’t agree more that relevance is the key to success. QR Codes are a great complement for both CURL’s and PURL’s. By encoding the relevant demographic information into the QR Code it can be passed to the site to drive delivery of the relevant messaging and imaging.

    This is all really about creating that dialogue with the client. And of course, once you engage them at your site you are able to capture additional information about their preferences which can be used to refine the future messaging and make it even more relevant.

  6. MichaelJ


    I was trying to figure out the path for a commercial printer to make CodeZ – information rich QR codes – part of their workflow. Please feel free to email me at josefowm(a) gmail.com if you think the info would be off topic in this interesting thread.

  7. Todd Thompson

    I agree with this post in some respects but there are two points I don’t believe are accurate given the direction many have been going in this wonderful economy.

    The use of a larger qty. when illustrating the the point of the less than cost effectiveness of a PURL campaign vs. a CURL campaign in many instances to me is invalid. We have seen our client base who used to do spray and pray direct mail now targeting their efforts and the effect is lower qtys. that are highly targeted thereby making the investment in PURLs affordable.

    The second is that all you are describing concerning a CURL is a versioned campaign which is and has been a common practice for GURL campaigns. Nothing new or earth shattering about it.

    The last item I’d like to bring up is that one thing is absent from this entire post and that is the value that a PURL campaign brings in the respect of gathering data. You mention the need for good data (which to our firm means relevant data) but leave out the fact that many do a PURL campaign as the means to collect new / additional information so they can start being relevant.

  8. Nick Pride

    Catching up late on a good post and an interesting conversation.

    One aspect which I think is a key part of the strength of personalised URLs is its ability to link marketing effort across multiple media (mail, email, online, even mobile and SMS). Create a relevant, targeted, campaign to a customer; present it in the right media (mail followed by email, for example) for that customer; use PURLs to create an equally targeted and relevant web experience; then we see campaigns that a) make more sense to the customer b) make better use of creative resources, and c) get better response rates. And we can even use PURLs to measure the impact of different media, and especially of the cumulative effect of different media – something that’s incredibly difficult to do otherwise.

    Add this to what everyone else has been discussing here, and the power is huge, and we’re seeing it working really well for our clients.

  9. Todd Thompson


    I agree with you on all counts and you mention very valuable points for others to note.

    Over the last year and a half or so we’ve been averaging 3 to 5 PURL or GURL campaigns per month. Many started as the vehicle to gather the data needed to be relevant other had sufficient data already but the win is that now our clients are enjoying response rates twice and three times that they were before they started this method. 5% to 8% is now common, double digits don’t happen every month but there are a handful of them.

  10. Michael Josefowicz

    I think you’ve raised the single most important point about Purls, Curls, dynamic QRs and whatever new technologies that will connect print to the web.

    You said “The last item I’d like to bring up is that one thing is absent from this entire post and that is the value that a PURL campaign brings in the respect of gathering data.”

    CMO’s don’t really need help on marketing campaigns. They have lots of people they manage full time to figure out the most effective way to do that. Printers actually don’t have a lot to add, in my humble opinion.

    However, CMO’s are under constant pressure to gather information about the behavior of their customers. The reality is that most of them don’t have a clue as to how to do it or what processes to use.

    The issue is not what the global can teach, it is what a global can learn. If printers joined with the best analytics companies we could help the globals create the predictive analytics that they will gladly pay for.

Comments are closed.