THE FIVE DEADLY HABITS OF FAILED
ANALYTICS PLATFORM SETUPS
The Five Deadly Habits of Failed Analytics Platform Setups and How To Avoid Them
The company had just made a major investment in a new marketing campaign. But was it working? That was, literally, the million-dollar question.
Company leaders rushed to launch an analytics platform that could track their key performance indicator (KPI): registrations. The initial numbers showed a surprisingly large jump. Success! They doubled down on the media buy. Then a sharp-eyed analyst noticed that something was off: every registration was being counted twice.
The whole point of investing in analytics is gaining the ability to focus on what matters most. This clear-eyed, data-driven insight, in turn, allows leaders to make better business decisions and avoid wasted effort and money. To get
there, you need to avoid these five deadly mistakes.
The Most Common Reasons Analytics Platform Setups Fail
In our experience working with clients like the aforementioned company, we see these five mistakes being made over and over.
1. Tracking Everything: Desperate for data — any data! — a company launches its analytics without figuring out requirements and business goals up front. The resulting analytics track every event on the site, driving up costs and confounding any actionable insights gained.
2. Failing to Launch: Let’s say a company wants to re-brand a core service offering, so it invests in analytics to measure its creative agency’s marketing effectiveness. Then someone high up changes the strategic direction of the re-branding, and the creative is scrapped — along with the analytics. The result: big spend, zero ROI.
3. Losing Faith in the Process: After migrating from one analytics platform to another, a company finds that the new numbers are drastically different from the old ones. It’s hard for the team to admit that they’ve been using bad data. And how do they know the new platform’s more accurate than the old one? Inconsistencies in the data lead the company to distrust everything the analytics are saying.
4. Backtracking for Miles: Once a team realizes its mistakes in analytics implementation, it scrambles to figure out what went wrong and fix the problems. This exercise can last an extremely long time and monopolize valuable resources that should be focused on new initiatives instead of historical errors.
5. Making Business Decisions Based on Faulty Logic: Bad logic can corrupt the entire analytics process and make it appear as if well-performing marketing assets are failing — or vice versa. Examples of faulty logic include events triggers that don’t fire consistently or an emphasis on last-click attribution, which fails to consider the end-to-end impact of all marketing tactics in driving sales.
Implementing a Flawless Analytics Platform
You can trace all these problems back to one basic error: failure to plan. Time and time again, we see clients (and even other analytics firms) skipping the discovery process entirely. This is a fatal mistake.
At Encima Group, we’ve developed a unique client discovery process that ensures a successful launch of every analytics platform. Here’s how we do things smart from the start.
1. To avoid data overload, ask the right questions. We begin discovery with a few simple business questions. What is important to you? What do you want to measure? How do you define success? We ask these questions not just once, but over and over again — to clients, their creative agency, their media agency, their analytics insights group and other stakeholders. We uncover a surprisingly broad range of expectations and goals. Our job is to align them.
2. To avoid wasted effort, keep your eye on the finish line. The main purpose of analytics implementation is to enable reporting analysts to quickly and accurately identify ways to optimize client marketing campaigns and websites to increase ROI. Goals shouldn’t be an afterthought.
3. To build trust in the process, make everything transparent from the beginning. Clients sometimes assume that launching an analytics platform is a service we perform for them. We see it as a collaborative effort with them. We want clients to understand exactly what we’re doing, so that in the future, they have the knowledge and confidence to lead the process.
4. To avoid backtracking, get everyone to agree on the route. When we come in to repair a faulty analytics program, we conduct a close examination of the existing implementation and gather stakeholder consensus on how it should have been done. Does the team agree on how triggers should fire? And how events should be tracked? Then we’re good to go.
5. To get the best data, align on standards. Using global naming conventions allows reporting analysts to spot trends quickly, spend less time scrubbing and understanding data, and spend more time looking for ways to improve marketing campaigns to drive efficiencies. In short, using consistent rules and methodologies makes everyone’s life easier and improves overall ROI.
Looking Ahead to Implementation and QA
We can’t emphasize the importance of planning enough. A thorough and thoughtful discovery process sets the stage for painless implementation and QA. The standardization of processes across client projects leads to cost and time efficiencies. For example: by standardizing QA browser and device options for one of our clients, we were able to decrease total time spent in QA by 30% across all projects in 2014. Our QA standardization efforts were so effective, the client implemented a similar process globally across all agency partners.
Our last piece of advice is to document everything as thoroughly as possible. As you move into the implementation phase, remember that documentation is key. The more detailed your technical specs, which are based on the requirements defined in discovery, the cleaner the build will be at go-live. The same detailed documents can be used to walk analysts through the implementation and show them how each request is tracked. And they can be consulted yet again to perform QA across a defined set of browsers and devices to ensure that the analytics enablement is functioning properly.
Remember the company that realized its analytics solution was double counting registrations? Here’s the scary part of the story: They’re the lucky ones. Nine times out of 10, when we begin working with a new client, they don’t even realize there’s a problem with their data.
We can help you diagnose the problem and align on a solution. Call 888-508-5688 or click the Contact Us button below for a complimentary, no-obligation Quick-Start Consultation. We’ll quickly analyze your current efforts at no charge and recommend an action plan.