Complete Customer Journeys & Attribution With Shopify Data | #204 Benji Walvoord

Please support the show if you liked today's episode:

• If you love the podcast, please get someone else to listen, too!
• If you enjoyed this episode of the Ecommerce Coffee Break podcast, please head over to Apple Podcasts, leave a rating, write a review, and subscribe.
• Share the podcast with your family, friends, and co-workers.
• Tag the podcast on Instagram @clauslauter and let me know what you like about it.
• If you like the content and would like to support the podcast, you can buy me a coffee here.
• Become a guest on the show or sponsor an episode.

Claus Lauter: Hello, and welcome to another episode of the E-Commerce Coffee Break podcast. Today we want to talk about data and about customer data and how you can attribute the data within your Shopify store specifically if you're growing. So in a nutshell, how can data help you understand your business and what you need to do to connect the right sources.

So therefore, I have Benji Walvoord and Eric Dodds with me on the show today, both of them are working for rudderstack.com and they are specialists when it comes to connecting data to the right sources, to the right software within your business stack. So let's welcome to the show. Hi Benji. Hi Eric. How are you today?

Benji Walvoord: wonderful. Thank you for having us on the show today. Doing great.

Claus Lauter: Benji, give me a bit background. Why is data so important for growing Shopify merchants?

Benji Walvoord: for a number of reasons. Shopify is a great platform for new e-commerce customers to jump on a platform and really focus on the products, their brands, their pricing, and let all of the really hard things about.

Setting up inventory, managing a website, managing transactions and security, letting Shopify handle all that for them. And as, as a platform, Shopify provides some very useful tools around, inventory management and even so far as product recommendations and that sort of thing.

Now data

Claus Lauter: is coming from all kinds of sites into a Shopify business. So you have all your apps collecting data, you have Shopify collecting data, you have your different sales channels, collecting data. It can become very confusing and obviously there is a ton of data and you wanna make sure that there is a bit of a stream in there.

Data goes in the right direction. I think that's why. Are you helping with Eric? All this data that's sources that are coming in and going out, what makes it so complicated to handle all of that?

Eric Dodds: The problem comes when your business starts to scale. So we have some very large enterprise customers running on Shopify, and like Benji said, Shopify has a great platform.

But let's think about a really simple example that becomes complicated as your business grows. And that's recommendations, right? Shopify has a great very basic out of the box recommendations system, but the reality is that companies are learning way more about their customers in different data sources that exist outside of Shopify.

So let's say you have a point of sale presence. let's say you have particular location data. Let's say you have a loyalty program that exists outside of the Shopify ecosystem. All of these are very critical for making the right recommendations to your customers.

And so it's not that Shopify doesn't provide tooling around that, it's just that the complexity of the data sources really outstrips Shopify's ability to deliver that to every customer, especially at the enterprise level. What we're saying is that in order to actually make the next step in optimizing your business, you really need to collect all this data in a single location.

The place that companies are doing this is in a data warehouse. And so the idea is that you get all of the data that Shopify provides you which is wonderful, but then you can combine that with transactional data. Point of sale data inventory data, even a lot of companies will have separate inventory data and it allows you to centralize it in one place so that you can actually begin to build a full picture of what's going on with your customer.

I

Claus Lauter: think that's a, the good keyword there that you said a one single picture of what's happening in your business and that's not the easiest task. No. You gave one example. Can you gimme a couple of other examples where you can basically from this data warehousing, where you can use that data in other applications in your business?

Benji Walvoord: That's one of the main utilities of Rodders stack is that not only does it allow you to easily collect the data, And send it to the warehouse. But rudders stack also standardizes the schemas of that data as it's being ingested. And so if you're trying to model data for business intelligence reports or powering dashboards, it's very helpful and it accelerates the effort by having that data in a uniform format.

Tackling business intelligence reports is the first wave of warehouse data adoption. The next phase of that is trying to understand a complete view of your customer. And so again, as Eric mentioned, you might have customer shopping behavior in that you're getting from Shopify.

You might have. Customer complaints coming from a phone system or a trouble ticketing system, and then shipping returns coming through some third party. And so if you want to understand as much as possible about that individual customer now you need to build that sort of golden record in the warehouse.

Claus Lauter: Makes total sense. Now you're working on, with big companies, with big brands like Allbirds, when in the customer or when in the company grows, is usually the point when they figure out is like it does not work with the onboard tools that Shopify provides. Is there is certain gross level when they come to you and the decision has been made?

Benji Walvoord: Yeah.

Eric Dodds: Benji, you wanna

Benji Walvoord: speak to that? It's not so much of a size as it is what teams are doing what with the data. And so a classic case is, we now have a different person or a different team working on our email nurturing versus our s m s nurturing campaigns. And now we have different teams creating similar campaigns or working on parallel initiatives.

But they're using different sets of data because either they, a, haven't instrumented the tools within Shopify the same way, or B, those downstream tools just don't support or collect the same sets of data. And that's why one of the things that we're, that we've, that we're very excited to have recently launched is our profiles tool.

And the Rudders Stack profiles tool actually creates a very straightforward, system or structure for creating user profiles and generating these customer 360 records. And so instead of creating many one-to-one integrations with integrations within the Shopify ecosystem our recommendation is to use Rudders stack to stream all of that data to your warehouse, build a fuller, better picture of the customer.

And then from your warehouse, send that data back to those cloud tools. And now all of your cloud tools are reading off the same sheet of music. Okay.

Claus Lauter: That's interesting. So you will have a full overview from all data sources within your ecosystem, whatever that is. And obviously most merchants deal with like 15 different dashboards, which can be very, very confusing.

And as you said, every dashboard only has a bit and a piece of the overall holistic view. And you bring all of this together. How work, how does that work from a technical perspective? Sure. Talk me through the process.

Eric Dodds: Yeah, I can give the high level overview. So one thing that's really unique about e-commerce businesses that's really a wonderful thing is that 80, maybe 90% of the intelligence that an e-commerce business needs is largely the same, right?

We have anonymous visitors coming to a website. Doing some sort of browsing activity, either making a purchase or not making a purchase. You have customers who have made a purchase and then all varieties of customers in terms of loyalty, lifetime value, repeat purchase, et cetera.

so what rudder Sack profiles does and why we're so excited about this launch was that it really removes almost all of the manual labor. Out of building those views with all of your customer data. Right. And so if you think about running campaigns against customers who have a total number of purchases or a total lifetime value of purchases or recent activity

when you have so many different data sources, it's very, very difficult, as you said, Klaus, to try to pull that together. What Profiles does is allows you to pull in the behavioral data from your website, the transactional data, and any other data sources that are relevant to you, and it will automatically generate all of these data points that allow you to immediately take action, right?

So I can see my highest lifetime value customers. I could cut that by margin, for example. That's a really big one, right? You may have high lifetime value customers, but if they return half their products you're losing margin. That's hugely important. So what Profiles does is allows data teams ingest this data and then get a model out of the box that gives them one row per user with all these really key features that represent the e-commerce funnel.

That allow their teams to take immediate action and then of course they can augment it and build on it over time.

Claus Lauter: Talking about integrations, every merchant has a different tech stack probably when they come to you. What kind of integrations, and I see on your website it's aton, but give me some couple of ideas as what does integrate with rather

Benji Walvoord: stack?

I would say Given the flexibility of Rudders Stack, there really aren't, anything that has a rest API that we can send data to, we can integrate with. And so it really are, we think about integrations as well. We have different mechanisms for ingesting data. So you might have an S D K that installs similar to a Google Analytics tag on in your webpage that can send real time events from the website.

We also have an e ETL product for extracting data out of CRMs and other cloud tools. And then, so then when we, in, when we ingest that data, not only does Rudders stack give you the ability to store all that data in the warehouse like we've discussed already, but there, there are also times where you'll wanna stream that data directly to some other cloud-based tool, whether that's a.

A tool like Braise or SendGrid or custom io or Attentive, not sure which of these are your sponsors, but we'll just say But , the beauty of Rudders Stack is that it does give you the ability to do both realtime streaming to these tools. So realtime, point to point integrations, but also to send that data to the warehouse so that you can enrich it.

Maybe filter it maybe aggregate events and to Eric's point build some sort of score that you then want to federate back out. But it really does allow you to make sure that all 5, 10, 15, 20 of these cloud tools or reporting endpoints have the exact same set of data. But universally, everyone has.

if they're using an app, they've got some sort of in-app messaging tool. They've got an email nurturing tool, they've got a web AB testing tool. There's SMS tools and of course some sort of Google analytics. That's a whole nother podcast in and of itself. And BI tools kinda that, that will set on top of that.

Claus Lauter: Okay. Now, how can you imagine working with the rather stack warehouse? What kind of front end do I have? What's the day-to-day view of what I would have in front of me as a merchant

Eric Dodds: For sure. So I can speak to that a little bit. We have a really nice UI and, and really we can think about is just making connections in a user interface Like Benji said, you install the rudder sack script that actually allows a l Most of our customers remove 80 or 90% of the tags that they're running on their sites because you can send data directly to those integrations in real time through rudder sack, as Benji mentioned.

So we see a lot of performance gains by running the rudders SDK as opposed to your 10 or 20 other tags. So you install that so you have a data stream coming in. If you want to pull data in from other sources like, Klaviyo or Stripe, et cetera, again, you just create those as sources and then you add your destinations.

And so you may wanna stream the realtime data to Klaviyo to trigger auto response email but you also want to send that behavioral data directly to your warehouse. And so it's just a matter of adding a warehouse destination like Snowflake. Adding your credentials and the data will start to flow into your warehouse with standardized schemas.

Very low technical lift. Although the product is built for data teams who want to support the marketing use cases and customer success use cases that the down downstream teams are trying to accomplish.

Benji Walvoord: Eric, you brought up a couple of interesting points that I'd like to double down on, which is, a lot of our customers come to us because they just wanna get data into the warehouse first and foremost.

But then, and particularly with Shopify, you don't have to install the S D K because our Shopify integration is native within the Shopify ecosystem. And you don't have to install anything. But one thing that customers, one of the reasons many customers buy Rudders Stack, Is that, as you mentioned, a lot of these customers will have 10, 15, 20 different integrations with Shopify, and what that means is that Shopify is going to install 20 different SDKs and fire, 20 different SDKs worth of network calls every time a page loads.

Or you click on a product or you view a collection. And a fundamental difference between Rudders stack is that with Rudders stack. You send the, you load one small s d K and you send one event to Rudders Stack and then Rudders Stack Federates all those events server side to all of your cloud tools. So it really improves site performance.

And then, just to clarify one more point, rudders Stack does not have its own warehouse. So Rudders Stack does not persist any data. So rudders stack, streams data to either your cloud tools. Or to your own warehouse, and it could be a snowflake, BigQuery, Redshift, Postgres, S3 Bucket, Databricks. Kinda we support a whole myriad, which you can find in our website.

Claus Lauter: now side speed. That speaks to me. Conversion rate optimization. I'm a big fan of that and obviously that's a big advantage there. Beside of having more transparent data coming in and out now, is there any kind of homework that immersion needs to do before he starts implementing? What kind of thought process do they need to go through before they approach you guys?

Eric Dodds: Yeah, that's a really good question. The biggest thing that I would encourage merchants to think about if you're on Shopify is number one do you have a data team who can support more advanced modeling of data in a data warehouse environment? And that's really where , our tool is built for to Benji's point, we don't store data we don't provide you with analytics and a user interface.

And so if you are considering moving towards a warehouse or moving towards a sort of warehouse based data strategy, or if you have already purchased a data warehouse and you are trying to operationalize it, get more value out of it rudder sack is really a great choice. I think site speed is another major consideration.

If you are trying to improve your site performance and remove tags and set up a one to many integration architecture from the rudders stack, s d k to all these downstream destinations, another great consideration. That usually falls within a data role or an operational role, maybe a marketing operations role.

But I think the biggest thing I would encourage people to think about if you. Are thinking about a warehouse or have a warehouse, and there are questions about your customer or your customer journey that you know will really help you reach that next level of growth, and you just don't have the data to answer it.

Or someone's exporting data from 10 different platforms and trying to combine it in a spreadsheet and spending multiple days to get a very simple answer. If you're in that place and a data warehouse is in the works, then you should definitely reach out. Because that's really where Rudders Stack can help you just drastically accelerate the time to value for unlocking those insights.

Betty, any other thoughts? And one

Benji Walvoord: telltale sign of that is if you have different user IDs or different customer identifiers that you're building. V lookups and Google sheets and crosswalks and all this stuff because all of these different SaaS tools now are some use an email, some will use a Shopify customer id.

Some customers say, we want to have our own internal id. That is the master ID for all these records, and it's hard to squeeze that in within the native Shopify ecosystem and the next step is, okay, let's get all this data into a warehouse and I have some basic SQL skills and I can tie all this stuff together.

That's usually a good indicator that it's time to start playing around with Rudders Stack we would certainly encourage folks to do so and have a free plan.

Claus Lauter: Okay. Not every merchant at this point has a data team. How far do you help with. Creating clarity on how to integrate Rudders stack into the tech stack over merchant.

Is that something

Benji Walvoord: you do? That's actually my principle role as a solutions engineer here at Rudders Stack. Spend most of my days helping customers answer some very basic questions like what is the definition of a customer? Is it an account holder? Is it a subscriber? Is it a purchaser within the last six months?

Or these sort of. Obvious questions over beers, but then when you start actually getting into the data, they become very complex and very meaningful because different teams have reported operational or success metrics based on a definition. And now that you're trying to get all this data in the warehouse and read off the same sheet of music, there can be conflicting definitions around something as basic as how many customers do we have?

Claus Lauter: Makes total sense to me. And again, I think it's creating clarity in the first place is a good starting point. And then going into the details and build up a system that can handle the data. Tell me a little bit about your pricing structure. How does that

Benji Walvoord: work?

Eric Dodds: We price primarily on event volume. And so you can think about user actions, or data points that are coming in. So that could be someone clicking something on your site or a stripe transaction being ingested into rudder sack. So we charge on incoming events or data points.

Is our primary pricing vector. We have a free plan so you can send, up to a million events per month for free. And then we have additional tiers beyond that. To Benji's point we have a lot of our large e-commerce cup, customers. Are on the enterprise tier.

Someone like Benji provides them support on not only figuring out their data model on the ingestion side and all that sort of stuff, but operationalizing our new product profiles, which can give them an e-commerce funnel and insights out of the box. And then also the activation of that through our reverse EGL l pipeline.

So we span the spectrum. We love encouraging people to go install the S D K. Stream some Shopify events or events from our JavaScript, s d k from your e-commerce website and connect a downstream tool and just try it. Is the best way to get a good idea of the platform.

Benji Walvoord: Yeah. And we've got a, we've got a Slack community that's very active and growing in, and our folks monitor as well. And you'll see Eric and I jump in there quite frequently.

Claus Lauter: Where can people find out more about Rudders Stack? Where's the best place to get to you guys?

Eric Dodds: Very easy.

Rudders stack.com. So that's R u d d e r, like about Rudders stack s t a c k, like a stack of pancakes.com. And if you scroll down on the homepage we have some e-commerce case studies that you can click through to see the types of results that our customers get. I will put the

Claus Lauter: links in the show notes as always.

Then you just one click away and I would highly recommend our listeners and viewers to give it a try and to figure out how they can optimize their data. Thanks so much for your time today.

Benji Walvoord: Thank you for having us. Thanks.