Analytics Tools for Omnichannel Experiences
How to instrument metrics for an offline-online customer Journey
The Snippet is a Weekly Newsletter on Product Management for aspiring product leaders.
Over the past few weeks, we’ve been talking metrics.
We talked about why having an End to End Analytics Strategy is key to driving growth for your product. We also talked about how to think about What Metrics to track at different stages of a product’s customer acquisition journey.
This is the final post in our “metrics” trilogy —and this one is about the Tools that you’ll need to actually instrument your measures.
As always, this will make much more sense when discussed against some context — so let me start by telling you a story.
Omnichannel Complications
Several years ago, I was working with my team to launch a Smart Home product in a highly contested category — Smart Thermostats. Our product was a Wi-Fi Thermostat along with iOS & Android mobile apps to control the thermostat.
Given our target customer segments and demographics, we landed on an omnichannel distribution plan, which means the thermostat could be purchased offline at major big-box retailers ( Home Depot, Lowe’s) and at regional brick and mortar distribution partners, as well as online - on our website as well as Amazon.
The Omnichannel strategy was to ensure that customers could find us wherever they liked to shop for their home improvement projects. And while this distribution strategy was working well, it also presented some pretty complicated customer journeys that were proving rather hard to measure and track.
For instance, here’s what a typical customer path to purchase would look like
A person looking to buy a smart thermostat searches the term “smart thermostat” or related keywords on google
They click one of our ads, or on the organic search results and arrive at our landing page
They browse through our product catalog and at this point can do one of three things to buy our product —
(a) They either buy a smart thermostat on our website right away ( tracking this purchase path is easy) OR,
(b) After having looked at the available options on our website, they leave our website and purchase the product at a nearby Home Depot or on Amazon ( tracking this purchase path is difficult) OR,
(c ) they simply decide not to buy it right away, but maybe sometime in the future (they become a lead)
Finally, — Once people bought the thermostat hardware, they download a FREE app and start using the thermostat via their phone.
I mention the above purchase routes to make a point — that the complexity of instrumenting and tracking metrics across the entire journey is closely tied to the number of routes that a customer can take to buy your product.
Your customer’s journey might start online (landing pages), then move to offline channels (Home Depot), and then back online (installing an app from the app store). In such cases, how do you track a lead-to -customer journey across all your properties both online and offline? What tools do you choose such that you can visualize the entire customer journey even though there were several offline events?
This why choosing the right tools upfront becomes so important.
TL;DR
This is going to be a long post, but here are the cliff notes on the tools we used to unify and track a customer across the entire buying path as they showed up online, dropped off our radar for an offline purchase of the thermostat hardware, and then showed up online again as soon as they started using the mobile app.
For Product & Landing Page Analytics — we tried using Google Analytics, but quickly moved away to HEAP analytics
For Marketing Automation & Analytics: Salesforce Marketing Cloud
For Lead Management Analytics: Salesforce Sales Cloud
For Customer Service & Retention Analytics: Salesforce Service Cloud
For Mobile Analytics: Appsee (acquired by ServiceNow, a good alternative is UXCAM)
If you are interested in learning more, What follows is a detailed explanation of why we chose these tools and how everything played together to track our Omnichannel customer experience.
#Product & Landing Pages
Google Analytics
There are several tools to get you started on your landing page analytics — but perhaps the most well known and most widely used is Google Analytics (or GA, as its often known as in the product marketing analytics world)
The best thing about GA is that if you just want a few basic “out of the box” metrics, GA is super fast to set up, and you don’t need to be a sophisticated programmer to set it up. Although you will need write access to the landing page HTML code to add the GA’s tracking code.
Here’s a quick intro 👇
The not so great aspect of GA is that setting up advanced tracking can get quite involved and you’ll almost certainly need help if you don’t know what you’re doing. This is what we kept running into.
If you are simply building a Side project and need basic analytics only — GA is perfectly fine and easy to implement. But if you have an involved product with a complicated path to purchase (online/offline distribution), implementing GA can add significant overhead.
For example, every button click that you want to measure on your landing page needs to be custom created in Google Analytics. If you forgot to create an event for a UI element interaction, all historical user interaction with that UI element is lost.
We needed something that was built for purpose. And that’s when I came across Heap analytics.
Heap Analytics
Heap is GA on steroids, and it’s much more builder/marketer-friendly than GA. It has a free version up to a limited number of sessions tracked and then you can upgrade to a Team or a business version.
Heap turned out to be just what we were looking for. It’s an analytics tool that combines Product Metrics, Marketing metrics, and Customer support metrics for your product and makes them available all in one platform —avoiding Frankenstein-ing different analytics tools. But the most helpful aspect of HEAP is that it automatically tracks your UI elements and makes interaction data available retrospectively.
This capability affords the Product Manager a great deal of flexibility in understanding user behavior, running experiments, and answer important questions quickly such as but not limited to — What is the typical behavioral flow? Who are your power users? Which activities predict long-term retention? What were users doing right before they dropped off?
Heap is especially suited for omnichannel models, as you can connect it to offline Point Of Sale data sources to complete your lead to customer conversion journey. For us, all our offline sales & customer activation data was in Salesforce and we connected that data to Heap to visualize the lifecycle of a lead that became our customer through one of our offline channels.
Here’s a handy table for comparing Heap vs other analytics options that you can check out.
#Marketing, Lead Management & After Sales
Salesforce
The vast majority of visitors that land on your website will not convert into customers right away. Some of them will instead want more information on the product and will leave you with their information through the Call To Action form on your landing page. A lot of them would bounce, but you’ll be able to retarget them based on web cookies. Bottom line is — these are people interested in your product (aka leads ) and your goal is to make them buy your product at some point.
Now, you need to store whatever information you know of these leads into a tool that then lets you market to them and convert them into customers. This is where Marketing Automation systems, Lead Management systems, and Customer Relationship Management (CRM)systems come in.
Salesforce is one of the biggest and most widely used out there. It’s one integrated platform that brings all of these systems together to provide a single, shared view of every lead and every customer. This is also what we used for our product.
For our Smart Home product, we used Salesforce’s Marketing Cloud to manage each lead, create targeted marketing campaigns for them. Once our leads converted successfully, we used Salesforce’s CRM functionality for customer support, customer communication, and retention programs. By the way, Salesforce is not the only option here — in-fact there are plenty others (e.g. Hubspot). We chose salesforce due to its scaling capabilities and a tremendous collection of useful 3rd party apps on its AppExchange ecosystem.
#Mobile App Analytics
Appsee/uxcam
As you’d imagine, the mobile app is a central piece in the UX and brand promise of any smart device. That’s where the smarts are! Naturally, we needed our mobile apps to work seamlessly for users at all times so they’d rate us highly on the app store.
We needed real-time analytics on how people were using our apps, which buttons were being pressed and where the friction points were. We also wanted functionality that’d let us access recorded sessions for ‘app crashes’ for quick troubleshooting.
At the time we used this great analytics tool called APPSEE for our app analytics but last I checked they’ve been acquired by ServiceNow and have closed shop since.
Another very similar and equally good alternative is uxcam. I checked their functionality and it’s indeed very similar to what Appsee offered. There is a detailed article on appsee vs uxcam functional parity on the latter’s website too!
#How Everything Worked together
All in all, we employed 3 tools to track our end to end, omnichannel customer journeys. Here’s a quick recap of how everything fit together in our analytics plan.
Our landing page had Heap analytics integrated so it would automatically track visitors to the page, which buttons are being clicked etc.
Our landing page had a Call To Action (CTA) form (powered by Salesforce) which served as the gateway for the visitor to become a lead. The CTA form submission created a lead entry into Salesforce, and we’d then be able to start targeting the lead with automated marketing campaigns.
Once our lead converted and created a customer account with us, we use Salesforce’s CRM platform to manage them, retain them as well as provide after-sales support. Everything we did with this customer could be tracked and measured.
Our apps had Appsee integrated to track and measure user experience, app usage, and troubleshooting purposes.
Finally and most importantly, For customers that converted via offline channels, Heap would be able to identify and merge matching past visitor records based on a common identifier (hardware id, email, browser cookies, etc). This would create a seamless view of the customer.
This has been a long post, but I hope this provides you with a handle on how to think about instrumenting metrics for complicated paths to purchases.
Thanks for reading. If you found value in this post, how about share it with a friend?
Want to connect? Say hello on twitter!
The Snippet is a Weekly Newsletter on Product Management for aspiring product leaders.