At first glance, building your own revenue analytics system might seem appealing. You get full control and can tailor everything to your specific needs. But is it always worth it?
Consider building if:
Building your own basic solution in house is possible with these steps:
1. Map out all your tools and KPIs
Step 1: Map out your tech stack and align on KPIs with the relevant stakeholders.
Step 2: Determine what’s really important for your business. Typical questions to understand this are things like:
Step 3: Determine what metrics each platform needs to bring to your Master Dashboard. Ensure you know where revenue metrics, form submissions, and cost will come from.
2. Bring all your data into a central spreadsheet or BI tool (aka your Master Dashboard)
At this stage, BI tools and spreadsheets are a great way to get started and get some visibility.
3. Build your first attribution models starting with First/Last touch
There are limitations on what you will be able to understand in doing this, but it is still the best way to start if you don’t have anything else set up.
Once you build these two attribution models, you can look at brand awareness channels/campaigns with first touch and conversion points with last touch.
4. Calculate your Blended CAC
Import your advertising, event spend, tool spend, headcount spend data into the same spreadsheet or BI tool.
From there, take the spend for a given period and divide it by the last touch attributed or first touch attributed number.
This will give you a result that is either weighted towards the very top of the funnel or the very bottom — but it is better than having no cost discipline.
Considerations during your preparation phase:
Reminders:
Now let’s talk about buying a pre-built solution like HockeyStack. What’s the upside?
HockeyStack takes a lot of the pain out of the equation by offering a no-code, plug-and-play platform. What sets it apart is its ability to help you not just answer what’s happening, but also why it’s happening. This isn’t just a data aggregation tool—it’s a way to make strategic decisions.
Here’s why buying might be better:
Much of the buyer journey is done through anonymous web activity, things like:
Internal builds can only track events that led to conversion (form fills). But as mentioned above, many touchpoints are anonymous website visits and other engagements from users who have yet to convert on the website.
In addition, most builds can only drill one level down into the channel or campaign.
Teams that bought HockeyStack wanted to be able to see specific ads, content, and assets to understand how each was impacting pipeline.
Finally, website data can only be tracked in internal builds via Marketing Automation (no pre-conversion data) or Web Analytics tools like GA4, which don’t match to account and don’t show how website engagement ties to pipeline and revenue.
If you are building internally, because of the inability to track anonymous activity or tie pre-form fill activity to contacts, reporting will be based on First or Last Touch. These attribution models are great for a single lens (when, where and why do contacts convert), but they miss dozens to hundreds of other touchpoints that occur in the buyer journey before the first form fill.
More advanced setups have multi-touch models, but they are based on campaign association (if an SFDC campaign member is associated with the opportunity). Even with this limitation, to be usable for analysis, it still requires:
The most effective way to understand marketing’s impact is to track as much data as possible and use different measurement methods depending on the questions being asked. Teams should regularly use first-touch, last-touch, and multi-touch models, as well as lift and incrementality reports. Even if you built out a full set of multi-touch models, they are strictly correlational in nature, and do not demonstrate causality.
To make confident, data-backed decisions, this directional data needs to be verified by lift reports (cohort comparison) that the touchpoint or campaign does impact conversion.
Constructing these models internally requires dedicated engineering resources, specific requirements and the ability to build these models on top of your visualization tool.
Marketing teams should be able to access relevant data and spin up reports in real time to answer questions as they come up.
For example, teams should easily be able to slice and dice the data as they please based on:
Building internally will likely limit data exploration to technical users if the tool is code-based and doesn’t have a user-friendly UI. If your team will need to put in a request for minor or major changes, the overhead of the in-house tool is simply creating another bottleneck.
To gain insights quickly and efficiently, every member of your team should be able to create reports in real time.
Internal builds take a massive amount of resources to build and maintain (estimate 7-10 months), and even the best builds are missing most of the buyer journey. The main reasons it is so difficult to build and maintain internally are:
Many teams feel they can build a solution like this internally because of their experiences with traditional attribution tools. But traditional attribution tools are based on a similar data set to internal builds as many rely on post-conversion data, converted contacts, and don’t include impression or engagement data either.
To actually advance insights into what’s working and what’s not, marketing and sales teams should be able to outsource reporting infrastructure and spend time drawing conclusions.
There are some obvious questions to ask yourself before embarking on an in-house build:
If you’re working on tight timelines (which, let’s face it, most marketers are), buying HockeyStack could mean the difference between hitting your goals this quarter or missing them because you’re waiting on a solution that’s still in development.
If understanding the “why” behind your customer’s actions is key to your strategy, and if you want a solution that combines ease-of-use with depth of insights, then HockeyStack is the way to go. But if you’ve got a simpler need, rock-solid data, and plenty of engineering bandwidth, an in-house solution might be a fit.
Whatever path you choose, the key is making a decision that aligns with both your short-term needs and long-term vision.
What’s your next move—build or buy? If you’re leaning towards buy, book a demo with our team or play with the interactive demo.
What exactly is HockeyStack? How do we help marketers drive revenue more efficiently?
Jump in to uncover the ins and outs of Marketing Mix Modeling – from its mechanics to its ideal applications – with Drew.
.svg){kind=icon}
.svg)
