If you’re a web analytics professional at a large company, you may have to work with several ad agencies or other partners that are responsible for managing different aspects of your online marketing initiatives (display ads, paid search, SEO, social media, email, web design, etc.). If it’s challenging to manage web analytics across just internal teams, it can be even more demanding when you add external participants to the mix. If you want to become a truly data-driven organization, you can’t limit your web analytics governance efforts to just your internal teams. If your partners are strategic to your business, they need to be a part of your web analytics focus.

When my kids invite friends over to play, they are expected to abide by the same “house rules” as my children (e.g., pick up toys, no fighting, no bad words, etc.). As a parent, it’s hard enough to get my own kids to pick up their toys consistently, let alone getting their friends to do so as well. Friends that repeatedly don’t observe these rules are not welcome in our home, and our kids are encouraged to find new friends to play with. Partners which are not willing to abide by your web analytics program’s standards should be changed for those that will. You need to extend your web analytics program to include your partners in order to be effective as a data-driven organization.

Are you working with the right partners?

Before I get into some specific best practices for working more effectively with agencies, you first need to evaluate whether you’re working with the right partners. A data-driven partner will not bombard you with lots of meaningless metrics — they measure the performance of their advertising efforts based on appropriate metrics or KPIs. Some ad agencies are still uncomfortable with the “t” word (transparency) and filter the results. Whenever alternative metrics are used in place of the expected KPIs, you know something isn’t right. Your spider sense should tingle when you hear something like: “The conversion rate on the last campaign wasn’t as good as we hoped for, but the view-throughs were through the roof!”

You want to work with partners that embrace transparency like an overzealous Care Bear. These data-driven agencies are aware of your key business goals, understand your KPIs, and are closely measuring the success of each campaign against those key metrics. Along with increased transparency comes the client-agency understanding that some failures will occur along the journey. Without full transparency, partners and clients will not learn from these bumps in the road and can needlessly waste advertising dollars. Data-driven agencies learn, improve, and maximize your ad spend.

Data-driven partners also approach measurement considerations at the beginning of a project instead of rushing through them at the end. I don’t know how many times I’ve seen situations where a client’s agency is scrambling to add tagging the day before a website/campaign launches. Would you rather buy a new car from a manufacturer that approaches safety from the very first design of the car or one that scrambles to include safety features on the production line?

Best practices for ensuring web analytics success with partners

Many enterprise clients have asked our consulting team for best practices in working with multiple ad agencies and other partners in terms of web analytics governance.  In our experience, there are five key areas that can make a big difference in how successful our clients are with their agency partners. Even if you’re only working with one partner, many of these best practices still apply.

1. Training and Certification

  • Ensure all agency partners have been trained on Omniture products and are certified Omniture partners. The expertise of the agency web analytics staff can vary greatly from partner to partner, and Omniture certification can be a good yardstick.
  • Reduce your roster to a smaller group of certified partners. Work with your internal teams to see how the active list of agencies or partners can be consolidated.

2. Documentation

  • Require a Business Requirements Document (BRD) for all projects, which includes tagging needs. Most partner projects will include a BRD, but many times measurement requirements are not factored into the business requirements gathering and definition.
  • Provide a corporate standards document to partners so that they know what standards are in place and why they are important to overall measurement efforts.
  • Create a process guide with clear instructions on the entire process from beginning to end. This document can come in handy if you’re working with a new agency or need to formalize the process with an existing agency. It will help with establishing clear expectations for both sides.

3. Single Point of Contact

  • Designate an internal contact person who is familiar with the tools and standards to field all questions from agencies.
  • Require a single point of contact from the agency side as well. If there’s an issue with the tagging which needs to be corrected quickly, you need a designated go-to person.
  • Hold weekly or bi-weekly meetings to coordinate efforts across different agency projects.

4. Quality Assurance / Sign-off Process

  • Require agencies and internal IT teams to perform QA on all tagging – both from a technical perspective (e.g., data is being collected) and business perspective (e.g., data is sound). In addition, make sure that any new marketing initiatives (including testing and redirects) do not interfere with current site tags or implementations.
  • Build validation of implementation and reporting into the final sign-off process to ensure specified business requirements have been met.
  • Structure your partner contracts so that post-launch changes or enhancements are not painful to execute. Rather than waiting for a campaign or website to run its course before analyzing the results, it can be valuable to make optimizations mid-stream in order to drive greater overall success.

5. Accountability

  • Hold agencies accountable for non-compliance with standards or poor quality of tagging. Lost or bad data translates into missed insights and optimization opportunities, and partners need to be aware of the ramifications to the business.
  • Hold internal groups accountable for not following the established process for agency engagements. The partners may be willing to follow a defined process, but never learn about it from the rogue internal teams they are working for.
  • Create alerts to catch potential campaign or website tagging issues early. A thimble of good data is better than a bucket of bad data.

If most of your online marketing execution is handled by partners, your organization will need to build partner considerations into its web analytics program.You can’t ignore the fact that ad agencies and other partners can make or break your analytics success. In order to achieve your goal of becoming more data-driven as a marketing organization, you’ll need to work with capable, data-driven partners. Using the right partners and the highlighted best practices, you’ll be able to achieve new heights in web analytics awesomeness and more importantly greater success with your marketing efforts.

Brent Dykes
Brent Dykes

Richard, Thanks for your comments. As Mark Twain stated, there are three kinds of lies: lies, damned lies, and statistics. As you know, web data can often be "massaged" in any number of ways to show exactly what the audience wants to hear. Ideally, your trusted partner isn't going to do that as transparency is critical to really understanding how your online marketing efforts are performing and optimizing them. However, I have seen companies that have done what you proposed -- essentially creating a "separation of church and state" where an independent partner or group analyzes the campaign and website results. It is an option and is a role that our consulting team often performs. Brent.

Richard, Web Design Leeds
Richard, Web Design Leeds

Completely agree about transparency being very important. If companies can afford it, I'd probably actually recommend that they occasionally get someone with no interest in the results of a campaign to check that the data they have been given has not been cherry picked for a positive outcome, and that it has been correctly gathered. Certainly, many glowing SEO reports I have seen are not so hot when analysed.