Are you a CMO? When was the last time you invested money into Customer Acquisition? Today?
When was the last time you felt confident spending one dollar because you’d get back 10 dollars? And I’m not speaking about extrapolating LTV. I’m speaking about real LTV on a specific customer.
Tracking Customer Acquisition can solve all your problems. But it’s not as easy as it may sound. Many solutions are available but they all have pros and cons…
This article will give you all the solutions to help you measure Customer Acquisition and compute your Customer Acquisition Cost (CAC).
Why should we track Customer Acquisition?
Customer Acquisition often takes up most of your Marketing budget. It’s important to know how you’re doing to adapt your strategy.
It’s primary that you measure progress to avoid going off-course and metrics are a good proxy for progress.
Tracking Customer Acquisition allows for a better control of your spending and makes sure you’re being profitable on specific channels.
To give you a quick example, we started writing answers on Quora with one of my customers. We quickly noticed leads were pouring in. Awesome, right?
After carrying out more research, we realized these leads weren’t converting and Quora wasn’t worth the investment.
It’s easy to see leads come in, it’s a little harder to consider the full-funnel and see if these leads actually convert to paid customers (which is what really matters).
Easy question: Why would you put money in a channel that doesn’t show a positive ROI?
This article is all about telling you how to track that and the different solutions at your disposal.
Quick disclaimer: Branding effects being very hard to measure, I won’t get into much details about them. (e.g. It’s easy to see Quora driving a lot of eyeballs but if we can’t measure the impact, that probably shouldn’t be run by the Growth Team).
Define your Metrics
Your business is very different from others: You can’t always rely on “template metrics”.
To give you an example:
- HubSpot realized most sales were closing at the end of the month.
- Churn of the sales closing at the end of the month was higher.
- They started implementing the Customer Happiness Index (CHI)
By implementing the CHI score, HubSpot was able to align Marketing, Sales & Customer Success on one metric.
How valuable could it be to understand how different channels behave when considering the CHI score for HubSpot?
For the little story… HubSpot hypothesized that these sales were closed because sales people were using aggressive tactics to meet their quota.
The CHI score made sure sales people weren’t too pushy on their leads.
Back to business. You need to define your own metrics. To help you out, Myk Pono wrote a great piece on how to define your metrics to track Customer Acquisition.
From what I’ve noticed, most companies rely on first / last touch attribution.
Although they’re the simplest attribution models on earth, they fail to depict the entire customer journey.
Consider someone who has the following interactions:
- Read your blogpost after clicking on a Tweet
- Search on Google for ToFu keywords
- Click on a Google Ad with an intent keyword (e.g. buy X)
- Click on a Retargeting Ad
- Convert after clicking on a organic result with an intent keyword
With this scenario, Google AdWords will get absolutely no conversions. Does that mean you should stop Paid Advertising?
For the record, here is an image showing you how the main attribution models work:
The attribution model you choose should be tied to your objectives.
- If you focus on Awareness, choosing First Click can make sense.
- If you’re focused on Customer Acquisition (like most startups), you should probably go for a model that depicts the entire journey (linear, position-based).
What’s important is that you should be able to switch between attribution models. That could allow you to do analysis like:
Content is awesome for Brand Awareness but we need to do some Retargeting if we want to increase website conversion rates and improve our results.
If you want to learn more about Attribution Models, check out this article by Avinash Kaushik: Multi-Channel Attribution Modeling: The Good, Bad and Ugly Models.
Channels VS Interactions
It’s important to look at the Channels bringing traffic. It’s equally important to look at what people are doing on your website.
To give you an example, if you notice that Google Search is an awesome channel for you, where is that traffic landing?
It could come from your blog or branded search terms. In one case you’ll focus on getting more high-quality content out there, on the other you’ll focus on brand awareness.
Without diving a little bit deeper into the data, you won’t know how to impact it.
Things to Consider
While tracking Customer Acquisition may sound simple at first, here are just some of the things you’ll need to consider:
- Ability to clean the data where neessary
- Ability to cater for your own business needs (targeting Accounts VS Users)
- Ability to integrate with other software (e.g. CRM, Payment Processors)
- Ability to manage all the Ad Platforms you use
- Ability to monitor all costs (CAC calculation)
- Ability to choose and switch between Attribution Models
The Perfect Solution?
Please note that I won’t get into the details of how to track links through UTMs. I’ll focus solely on the stack that you need in order to correctly track Customer Acquisition at a SaaS company.
Having a Data Warehouse means having all of your data easily accessible from one database.
Having access to the raw data is awesome because you can do whatever you want with it: you aren’t limited by a specific software.
You’ll also need to push data from your entire stack into this database. Think about CRMs, Payment Processors, Customer Support…
Fortunately for you, we’re not in 1995 anymore; a few companies took the pain out of developing all these integrations.
Once that data is within one single location, you can either:
- Query it yourself and ask developers for custom dashboards (Make)
- Use a Business Intelligence tool to make sense of all that data (Buy)
Depending on your size, I’d probably encourage you to Buy and not to Make a BI tool for yourself. They often get bloated and lack attention because they take away ressources from the product side.
Although these tools can be expensive, if you’re doing $70k+ in MRR, they’re totally worth the cost given that they should help everyone take better data-driven decisions.
However, unless you have some developers with free time (that doesn’t happen, right?), you’ll most likely need an analyst to make sense of that data.
- Access to the raw data
- It’s automatic so you won’t have to spend hours working through the data
- Ability to choose and switch between Attribution models
- No historical data (depends on your provider; you could also push that data yourself)
- Hard to access: You need someone who can write SQL and make sense of that data (it’s not easy peasy)
- Limited integrations: if an integration isn’t provided, you’ll need to develop it yourself
- The entire stack can be a little expensive
On that last point, expensive doesn’t mean that it isn’t worth the investment. If you’re reluctant, start pushing data to a warehouse and invest in a BI tool later on when you feel the need.
Google Analytics is the most well known Analytics software on the planet.
It’s great to see what people are doing on your website but it can be hard to configure for SaaS businesses. This is mainly because the sales cycle is longer than the session duration.
If you’ve configured a “Paid Customer” Goal into GA, most of your conversions come from Direct or Email Traffic and not the initial channel that led to a conversion.
You could overcome that by:
- Tracking LTV into GA as a Custom Metric but it won’t be as flexible as having your Data Warehouse.
- Pushing Revenue and assign the User ID (reference)
Correctly setting up correctly GA on SaaS applications isn’t easy so I’ll probably write an article on that later on.
GA works great with Google AdWords. If you want to have other costs into your reporting, make sure to check out Funnel.io. It will allow you to bring data from multiple datasources (e.g. Facebook, Bing) directly into GA.
- Lots of options in terms of reporting
- Strong on attribution models
- Slice data in any way you want
- Set up can be complicated (you’ll most likely need an expert)
- You’ll need a few hours to understand how it really works
- Accuracy is hard to achieve (sampling)
HubSpot – the Inbound Marketing solution of reference – has some features that can help you track Customer Acquisition.
It’s quite flexible but you might have to do quite a lot of work to prepare your data and make it look meaningful. Doing this work is no easy task and requires working with Workflows and other tools.
This solution is awesome if you’re part of the ecosystem. Your website runs on HubSpot. You use HubSpot as a CRM. You have integrated it with your APIs. You’ll get great data.
On the other hand, if your website runs on WordPress and you use Pipedrive as a CRM, you may struggle to make everything work.
Finally, it’s an awesome tool but it’s also simplistic. After all, HubSpot isn’t an analytics company. When you eat and drink data, you will miss a few things.
- Easy to implement and use
- Integrated with all your Marketing tools
- Lacks flexibility
Excel / Manual Gathering
Open up an Excel or a Google drive and start gathering your metrics. It’s probably the simplest solution but also the most time-consuming.
It can be a good solution when you don’t have the ressources (mainly money & time) to invest into other solutions.
Pay close attention to the metrics you gather because each and every one of them will take you time.
I’ve seen people gather more than 30 metrics. They spent an entire day gathering that data every week. What a waste of time when you know that only 5-6 of them were actually useful and actionable.
- You can access all historical data
- You have a huge flexibility
- Impossible to switch between Attribution Models
- You have to spend a few hours per week / month actually gathering numbers
- Low granularity: since you have to gather all the metrics by yourself, you probably won’t gather the most granular metrics (e.g. Campaigns, Ad Content)
Having a Data Warehouse and a BI tool from day one is realistic, but is this really where you should be spending your time?
What happens in most businesses I know is that they’ll wait until data becomes a liability in order to start investing in it.
I wouldn’t advise you to wait until it’s a liability but you shouldn’t live in a huge house if it’s only you and your dog.
Know where you stand and choose the right Analytics Stack for your needs.