Tracking users across sites and across touchpoints is increasingly difficult. Between regulatory constraints and the increasing blocking and rapid expiration of third-party cookies, what is the marketer to do when it comes to determining the impact of their multi-channel (omnichannel!) marketing‽
At our February 2021 (virtual) meetup, data scientist Dr. Joe Sutherland from Search Discovery described both the challenges (and inherent shortcomings) of “traditional” multi-touch attribution and then put forth a radical, but, in many ways, not at all new alternative: randomized controlled trials (RCTs). In a nutshell, it’s like an A/B test…but using randomly assigned geographies (zip codes, DMAs, etc.) to “split” the media. This includes everything from a “basic blocking” design (straight up “on in some, off in others”) to “stepped wedge” designs (every geo gets some media, but the level and timing of the media varies).
This approach provides truly causal results, which is something that media mix models and traditional multi-touch attribution simply could never do! Plus…it doesn’t rely on user-level tracking (it’s based on where investment is made and the results of those investments). And there was a chihuahua muffin (it’s a machine learning thing):
At this month’s event, Kristyn Wilson from Upright presented to a packed (virtual) house on the topic of “trust.” It may sound like a squishy topic, but there has been a lot of research into the impact of trust on brands, as well as the shifting nature of trust as the customer journey has shifted online over the past two decades.
As Jack Ma, the founder of Alibaba, put it:
“When you trust, everything is simple. If you don’t trust, things get complicated.”
Or, if you’re more into hard numbers, Accenture conducted a study of 25,000 consumers and estimated that a lack of trust costs U.S. brands $756 billion per year! That’s because consumers switch brands or providers due to lack of trust.
Viewed through the ever-evolving and oft-reshaped “marketing funnel” lens, trust increases efficiency and throughput at each stage of the customer journey:
But, what are the actual components of “trustworthiness?” Kristyn referenced Rachel Botsman and noted that there are four distinct traits involved (these apply to brands and to individuals!):
Competence is about how capable the brand (or person) is when it comes to doing something, including delivering a good or a service
Reliability is about whether that expertise can be consistently delivered and depended upon
Integrity is all about honesty and fairness
Benevolence is all about goodwill
Just as the traits of trust are not unidimensional, the tactics for building and maintaining trust are not, either, and they can’t be ramped up and “done” overnight. Kristyn described the “Upright Indicator” as a five-component tool for measuring a brand’s (and its competitors) trustworthiness:
It was an informative and engaging session that included a great discussion about the role of trust in a crisis (tip: it helps to have established trustworthiness before the crisis, and then it helps to have transparency and integrity during the crisis!), how to build trust as a startup (Get out there and engage! Build up that competence!), and even a discussion of Jeni’s Ice Cream!
This was our last event of 2020, but we expect to be back (virtually) in 2021!
At this month’s (virtual) meetup, Grace Koplow and Cory Underwood from Search Discovery provided an overview of the myriad movements in the world of browsers that are impacting the world of digital marketing, including analytics, optimization, and digital media.
For the attendees who arrived thinking that these impacts would, surely, be a mixed bag of “for the better” and “for the worse,” well…it’s complicated. From the perspective of consumers, there’s a lot of, “Oh…this is better! I’m not being digitally followed hither and yon without really realizing it!” From the perspective of the marketer, there’s a lot of, “Wait. You mean it’s getting harder to follow consumers hither and yon digitally?”
At its most basic level, there are two fundamentally different forces acting on marketers ability to track their customers digitally:
Regulatory Forces—this is the world of GDPR, CCPA, the almost certainly soon-to-be Prop 24, and many, many, many more. This is the territory where an organization’s legal department gets involved, opt-in / consent solutions get implemented on the website, and analysts and marketers have to really start caring about what data they’re collecting and how they’re managing it. This was not the main thrust of the event, but it was important framing at the outset.
Technology (Browser) Updates—while regulations require companies to take action to be compliant with regulations, companies have no control over the browsers their customers and prospects use when visiting their sites. And, increasingly, browsers are cutting back on what they allow cookies to do when it comes to tracking users across sessions and across sites.
From the past and future (near-term and medium-term) changes, to the many ways those changes will impact organizations, to an active discussion with attendees about strategies and tactics for changing how they work, it was a lively event!
One of the questions that sparked some discussion was, “Can enough smart people get together and try to figure out a long-term solution that works for all parties?” Cory noted that such a group actually exists as the Privacy Community Group on Github. Some other handy resources on this topic that keep track of the nuts and bolts of this topic:
If you’re interested in getting the full details of the event, check out the recording and/or the transcript below:
Grace Koplow: So yeah, today…whoever is here because they read the title and thought this was cooking related. I’m gonna just start off that that I’m sorry to disappoint.
Grace Koplow: I figured it. That’s why we had such a high turnout. So I understand if you do want to draw. It’s fine. It’s not about the taste of desert and unfortunately we’re not in Ohio. So Tim can provide tasty desserts.
Grace Koplow: But for the rest of you, we can talk about the ever so exciting browser updates and death. The cookies.
Grace Koplow: So today we’re going to talk about kind of start with the foundation of these updates the bread and then jump into the browser updates specifically potential features and impacts possible solutions and really what you should be doing right now.
Grace Koplow: So what are we talking about cookies used to be the company’s best friend. It was the little nice identifier that help companies know all about their consumers were
Grace Koplow: Been what they’ve been looking at how to predict what they might do essentially pretty foundational elements to a company
Grace Koplow: Well, today, we’re seeing large efforts to limit and in some cases stuff all the other tracking and identification methods now with these methods of identification on place comes a rightful concern for people for their privacy and the need to regulate what kind of information is being
Grace Koplow: So that brings me to kind of what regulation means and worse than regulation in two different ways. One, which most of you probably have heard about, which is the legal side of things.
Grace Koplow: This is the government regulation being put in place to ensure consumers are aware of companies tracking information and gathering and provide that opt in, opt out functionality.
Grace Koplow: To kill people have transparency into what personal information is being collected about them.
Grace Koplow: This is those GDPR CCP PA and for those that are keeping up with the trends is EPA has a new book coming out in November, that can make it even more stringent.
Grace Koplow: So there are a lot of discussions and updates around the legal side of things that are being put in place to help from a privacy perspective.
Grace Koplow: Addition of those. There’s the industry responses. And that’s the part that we’re going to talk about today and
Grace Koplow: From an industry side. I mean browsers like Safari or Chrome or Edge or Firefox and even iOS.
Grace Koplow: Now they’re taking this privacy initiatives into their own hands, a bit and updating settings on their front.
Grace Koplow: To limit data and information tactics all together without the opt in, opt out features just automatically kind of cutting data collection availability off at the source.
Grace Koplow: Just so we’ll be diving into that more about what that is and the impact during this presentation, but ultimately here wanting to distinguish between the privacy laws that are being put in place and the industry responses to what data is being collected and how it’s being used on Israel.
Grace Koplow: You might be thinking to yourself, Well, I’ve heard of Jesus and CC or GDPR excuse me and Susie pig, but the industry stuff sounds like a pretty big deal to
Grace Koplow: Well, it is. And things have been happening over the last couple of years, but you might not have been impacted or seen too much.
Grace Koplow: aspects to the data, your data been changed for it’s been on a more browser specific level.
Grace Koplow: As you can see, over the last even just this past year and Safari, Firefox, have all been making little tweaks to their data privacy and being able to restrict it at different levels.
Grace Koplow: So Safari here has been the one taking the main aspects of the main lead to this, but so far is about 16% of the US market, which is why you might not have heard of it until today, or just recently.
Grace Koplow: And in recent times why probably most of you are here is the September Apple released their IOs 14. A lot of us have that phone had that silly update that we always have to do that breaks our phone for a little bit and all that fun stuff. But what it did was it
Grace Koplow: Essentially impacted all browsers come in from iOS, which is unfortunately 53% of market share versus that little 16% of web market share that Safari had so you might be starting to see it impact if you haven’t already from these updates.
Grace Koplow: And didn’t see a question. What are your thoughts on the theory that Google and Apple or do is to set up their own DSP inventory
Grace Koplow: Or Korea. Are you from the trends side of things. If you have thoughts on where you think it’s going or the root of all this.
Cory Underwood: I think that
Cory Underwood: It would be fair.
Tim Wilson: To say that again.
Cory Underwood: I think it’s more likely for Google than it would be for Apple, because
Cory Underwood: Apple is pushing for a privacy friendly advertising network.
Grace Koplow: As much as it’s they’re calling it privacy friendly. It’s a more of a pain to the advertisers and the workers. So take it for what you will, um,
Grace Koplow: But a lot of it. You know, it really is rooted in this big push of trying to just help limit what data is being collected from people
Grace Koplow: And how it’s being used and that’s going to kind of continue to be seen, you know, from long timeline made short things have been unfolding a bit more with rapid pace, which we see
Grace Koplow: In start bringing to people’s attention now because of these recent updates with Apple, in particular, and what we continue to see over the next few months. And next couple of years, or really kind of snowballs.
Grace Koplow: And at the end of the day, the industry updates that are unfolding are affecting every online businesses ability to consistently identify visitors, leading to sub optimize marketing spend
Grace Koplow: Incorrect personalization and eliminate a B testing advocacy. Now I that’s a extremely bold statement that I just made about something that from our poll. Some of you might be hearing for the very first time. So let’s kind of double click a little bit into that and understand
Grace Koplow: Sorry, I double click a little too far. Give me one second.
Grace Koplow: My whole presentation got so excited about browser privacy updates.
Grace Koplow: At this is kind of
Grace Koplow: What happened. Let’s try. Yeah.
Grace Koplow: Double clicking only this far into things.
Grace Koplow: And talk about the browser updates and really how it is impacting you
Grace Koplow: So let’s talk about what actually is happening. Have any you’ll be seeing 51st dates.
Grace Koplow: For those of you that have. It’s a great movie, let’s all agree. And for those of you that Herrmann essentially Drew Barrymore had an accident.
Grace Koplow: Is given her short term memory loss and she couldn’t remember anything past the day
Grace Koplow: And good old cookie Adam Sandler began to a lot of Drew Barrymore and would have to recreate the first day every day since she kept forgetting. Now imagine that you’re a company who’s like Drew Barrymore and anytime someone came to interact with you, you forgot everything after a day.
Grace Koplow: That’s essentially the thoughts that are
Of the browser.
Grace Koplow: Browser or blogging or refreshing data anywhere from 24 hours to static
Grace Koplow: So knowing this, not having that knowledge can affect a wide range of elements for a business from audience segmentation is becoming inaccurate retargeting optimization functionality destroyed it all sorts of things.
Grace Koplow: Are starting to be impacted because we don’t know who these consumers are and how they’re interacting with sites.
Grace Koplow: And by not being able to understand. Remember, and interact properly with your consumers, it really boils down to the fact that these updates prevent companies from knowing who consumers are
Grace Koplow: Where they come from what they did on your site and did anything that you did matter.
Grace Koplow: Like your personalization or marketing efforts that you’re throwing billions of dollars into because once people left for the day, or that week, you might not have any data anymore. To give you these answers.
Grace Koplow: So I’ll pause here for a little second and let that question thinking is, can you imagine having to be Drew Barrymore and figure out why Adam Sandler is coming up to you every day with no recollection
Grace Koplow: Now, I love a good Adam Sandler movie as much as the next guy, but that sounds a little bit torture us
Grace Koplow: As well it’s happening to businesses now and as businesses not understanding that these core functions of your consumers is going to affect your marketing optimization your analytics and your business decisions.
Grace Koplow: Now we kind of get an understanding of what just happened or what’s happening and going to continue to kind of snowball, let’s talk about why you really want to start caring about it and how it’s impacting your business.
Grace Koplow: Overall marketing perspective, you know. Oh, there’s a wide range of this could kind of be impacting anywhere from your ad spend and accuracy broken attribution models.
Grace Koplow: Inability to remarket consumer journey mapping is limited, a whole bunch of things that efforts that you’re doing marketing. A good way to example of this.
Grace Koplow: Take a second to think you’re running a Facebook campaign and someone clicks on your ad like me.
Grace Koplow: And I go to the site and I really like what you purchase it or what you’re selling and I put it in my cart.
Grace Koplow: And dinner’s ready. So I leave my phone and I forget about it. And there’s some shiny. He’s over here and a squirrel over there and I
Grace Koplow: Think this whole thing completely and eight days later I remember that I actually really wanted that thing and inevitably do what we all would have done anyway in
Grace Koplow: The company sees that as a completely new user and someone that has organically found the site and purchase on their own. They have no connection back to that ads.
Grace Koplow: So that’s going to really start impacting you know an example of how this is really affecting from a marketing person.
Grace Koplow: From an optimization side anything from a personalization becomes difficult loss of sticky treatments and underlying assumptions with statistical tests are starting to be violated an inaccurate.
Grace Koplow: Know, another good example is running an A, B test for 30 days because let’s be real, you need that amount of days to get a significant amount of traffic to make it a viable path.
Grace Koplow: Well, someone 30 days is way past that 24 to seven day window, we’ve been discussing.
Grace Koplow: That same person is going to potentially be in both buckets and have a much harder time authenticating your traffic and have a poor user experience because now the button is in
Grace Koplow: Two different places on the site because that’s where you’ve been testing it, and all these different things are impacting how you were able to optimize and really see what’s affecting your site from not
Grace Koplow: From an analytic side of this one might be the cool obvious one, because if it’s impacting data is block. Well of course then data is affected. But what does that actually mean data is affected.
Grace Koplow: Well, it’s affecting everything you need to know from users consumer journey to segmentation to
Grace Koplow: Inflation of new users wrong reason sees any kind of core foundational data that you’re basing your company off of in your decisions off of that you’ve been working so hard to get to is starting to be inaccurate last skew.
Grace Koplow: And so with that was your core foundations of analytics marketing and optimization being inaccurate.
Grace Koplow: That’s going to lead your business decisions anywhere from your financial consumer or product decisions that you’re making also these now skewed foundations can cause impromptu and inaccurate businesses in
Grace Koplow: The world who built your general foundation off of is changing.
Grace Koplow: And so we really need to make sure that we’re understanding how it’s being affected before you start pulling the different levers of your company because you’re seeing differences in a data might not actually be the data might be what’s actually happening with the best
Grace Koplow: Way you thought that was all doom and gloom toe Ori there’s plenty more.
Grace Koplow: These invites are expected to expand to Chrome by 2022 that little 16% that we talked about earlier from safaris percent of web market share. Imagine once it starts to expand to Chrome and all the other browsers.
Grace Koplow: Additionally, Google and Safari had proposed changes and how attribution works. So even within the space. Things are starting to change in that aspect.
Grace Koplow: Too far also wants to delete all client side stories. So when the user isn’t even log when users aren’t logged in having that information deleted as well.
Grace Koplow: And then, of course, any efforts that you’re trying to make like fingerprinting are seen in those things are being recognized and blocked as well or projected to be blocked.
Grace Koplow: Now ultimately the thing that I want you to take away from this past doom and gloom sections.
Grace Koplow: Is that you may not have heard of these changes or you might not be affected significantly over the past couple of years.
Grace Koplow: But things are snowballing quickly and presented quite the challenges as they do for your tools, your current solutions and your larger business decisions moving forward.
Grace Koplow: Don’t worry, I’m not here to just tell you all the bad things. There are approaches to mitigate these impacts and changes. So some
Grace Koplow: solutions that are starting to take in place, you know, is shifting a first party data and below is less technical folks like myself. It’s a fancy way of saying changing where the data is going and coming from to be from your site and have more control.
Grace Koplow: And
Grace Koplow: Additionally, you might start seeing more communication and and data trading between vendors that communication and partnerships within the industry may be one of the ways is map out a map your consumer data is having those sorts of relationships built
Grace Koplow: And finally kind of that third grouping is building models and or different states in which the data is routed to really help fill the gaps or provide the direction information as business decisions in
Grace Koplow: And so those three really kind of
Grace Koplow: Go down to those three buckets of industry solutions. Again, that’s the partnerships we were talking about of the having data going between different vendors
Grace Koplow: Vendor solutions which is more technical. You know, there are options on, you know, Google animators aliens all those more technical aspects to try and work around the last data.
Grace Koplow: And your agency internal solution. So, what we are seeing is a lot more leaning into modeling. So to model your data to help fill the gaps that are lost by these
Grace Koplow: If all those don’t work, or we’re not quite there yet. You can also take some precautions up front, by considering putting in a process things to validate your data quality data quality regularly so you know there isn’t something that’s there right now to do
Grace Koplow: It’s keeping up and making sure that you’re monitoring and not pulling those letters that we said from a business decision, just based off reaction of data are really making sure that you’re keeping up with what the processes and validating that quality.
Grace Koplow: Because unfortunately there is no silver bullet right now. And so that is going to most likely be a combination of those three efforts to try and figure out what you can do now and in the future in order to help work with the industry as it does continue
Grace Koplow: So now that I’ve told you that there is a problem. It’s growing and there’s no silver bullet, bullet
Grace Koplow: And I laugh a little bit because coriander, would my partner in crime here. His nickname at our company is Dr. Doom, because all of this sort of stuff is just sort of a little bit doom and gloom.
Grace Koplow: What can you be doing right now. Well, the big one is to educate, educate about your company and your impact.
Grace Koplow: How are you being impacted what is happening, what is being affected. What are those elements that you need to be aware of for the future impact.
Grace Koplow: And that’s other part of the education, which is staying up to date on what’s changed. It’s not a set it and forget it kind of
Grace Koplow: It’s quite the contrary. You, if anything, especially burnout affected and making any changes in the current situation, you need to keep up with the trends to me for those things aren’t snowballing and effectively.
Grace Koplow: From our foundation of education, that’s when you can dive into discussing options and strategies, you know, as I said before, because there is no real one silver bullet with all this
Grace Koplow: It may need to figure out what are the best options for the short term, long term. And what do we need to rethink all together.
Grace Koplow: And and pot and we can pause a great question of, you know, will Google and Facebook ads be considered first or third party.
Grace Koplow: Correct. And correct me if I’m wrong, but those are going to be third party.
Cory Underwood: Pens on how they work.
Cory Underwood: So if you want them to be like retargeting a specific audience on websites, other than the website that you’re loading their script on
Cory Underwood: Then they’re executing in third party and that does not work on iOS, it does not work in Safari, it’s probably not going to work as you think it would in Edge or Firefox.
Cory Underwood: If, however, you are only caring about measurement on your site specifically
Cory Underwood: Then you can run them in first party. And then you have to deal with the fact that you still might not be able to get data from edge in Firefox, but Safari could arbitrarily limit you do only doing data collection for 24 hours, which, by extension, applies to anything on iOS.
Grace Koplow: And
Grace Koplow: That answer of it depends, unfortunately, is a lot of times the answer. And this browser privacy things
Grace Koplow: But also how come each company sets up the data on their side and tagging and tracking and all that sort of stuff. And Barry of how you’re actually impacted so
Grace Koplow: There is no again that goes back to, there is no one size fits all. But why it’s so important. Right now the educate understand how you are affected because it is so unique each client.
Grace Koplow: To be able to then make those strategies to be able to figure out how to navigate this in the future work around the future.
Grace Koplow: And so it after all this, you’re like, well, this doesn’t affect me. It doesn’t bother me. I don’t care. I’m sure I’m fine.
Grace Koplow: Who’s the schmuck talking and web analytics Wednesday. I’m sure she doesn’t know anything
Grace Koplow: Well, fine. I’ll take that for what it is.
Grace Koplow: But there are some things you could do to go check it out on your own. Just to get a gut check and this is by no means everything to check and by no means telling you exactly what’s wrong, because as I said before, everything is a little bit different.
Grace Koplow: But a good gut check to just see how you may be impacted is some of these areas to look in in for yourself. So the first one is using iOS, as far as traffic.
Grace Koplow: geyser said before this is the main areas being effective right now. So by seeing how much percent that you’re affected can really showcase how much of your data is being
Grace Koplow: Additionally, from your paid media side. You can check how your paid media is set up. If it’s deployed client side and time manager, like a GPS. It’s very possible that your pixels may be blocked or not working.
Grace Koplow: If you run an optimization. Check your testing platforms and see how your cookies are set how they’re set can show if you have blockers in your ability to persist experiences or have skewed results.
Grace Koplow: And then finally, if you want to jump into your code, you can look for a document doc cookie can show anything attached to business, most likely going to be effective.
Grace Koplow: And you don’t want to do any of that or again for you, less technical folks like myself, I will make a little plug here to search discovery.
Grace Koplow: We are offering free high level scans that do this sort of higher
Grace Koplow: audit of your current setup to see where you may be impacting to just help you start having those conversations
Grace Koplow: And just have you start understanding where you may be impacted and just start looking into where those things are. This is really a big place to start. And it again worth in education.
Grace Koplow: And
Grace Koplow: Pause for a good question of what do we think about the proposed hash email method for identifying users.
Cory Underwood: It’s largely going to depend on how that technically works but additionally to that.
Cory Underwood: There are legal aspects for using email for marketing purposes. So depending on the jurisdiction of both where you are and where the user is that may not even be viable because legally, it would be prohibited so
Cory Underwood: You have to figure out, can you legally do it. And then can you technically do it and then does the customer actually provide the email to make it work.
Cory Underwood: So there’s a couple of different scenarios, you have to walk through to think about the likelihood. With that said, we are seeing a lot of retargeting vendors, use a hash email or log something common between two sides to power that remarketing
Cory Underwood: Now the question is going to be
Cory Underwood: Should prop 24 for California pass that might be illegal inside of two years so
Cory Underwood: The next couple weeks are going to be very interesting.
Cory Underwood: Because if California is law passes, then
Cory Underwood: We are going to see a large shift in how what they term as cross
Cory Underwood: Textual consent behavioral advertising takes place so correct cross context behavioral targeting
Tim Wilson: Can you say that three times really quickly.
Cory Underwood: You can use it personally identifiable information such as email for marketing purposes, except in cross context behavioral advertising.
Cory Underwood: Targeting because that is different. So they actually want you to get consent prior to using it in a retargeting scenario, even if you own both of the websites as long as they are distinctly branded it’s considered cross context.
Cory Underwood: So we’ll see if that passes.
Cory Underwood: Current polling has a 77% chance of passing
Grace Koplow: And this all
Grace Koplow: Goes back to our all the things happening to regulate and this one area that we’re diving in allows us
Grace Koplow: I want to be careful how I say this flexibility and try to work around the system because it isn’t legal regulated.
Grace Koplow: So that you know you have your legal privacy where we can help you comply, because we will help you comply
Grace Koplow: Or you have your industry ones, which I think we’re all trying to figure out a little bit more of how we can help recover or figure out ways around getting the data needed that may have been blocked.
Grace Koplow: And
Grace Koplow: So real quick, just sort of the last thing, and then I would love to spend even more time on questions.
Grace Koplow: Key Takeaways if there is nothing else you leave today with other than this was not a cooking show browser updates are happening, and they will continue to impact your data.
Grace Koplow: This, this will cause gaps in your company’s data effective marketing optimization analytics and business decisions.
Grace Koplow: And a really, really good next step, as we’ve already kind of gotten a hint into is because everything is so unique to each situation start to educate yourself on how this is affecting and going to affect your business to ensure successful execution.
Tim Wilson: There’s another question, but it does back into GDPR what you things you can’t do I believe
Cory Underwood: So even if you get permission from the user. If the browser blocks. It doesn’t matter.
Cory Underwood: So we’re not only seeing things that you need to implement from a legal framework point of view, where you need to get consent party using data in this way.
Cory Underwood: However, because of what the browser’s are doing. We’re also forcing a heavy retooling of how websites functionally work under the hood. So a lot of the client side tagging and stuff that became popular over the past decade.
Cory Underwood: Is subject to say ad blockers and that used to be. Okay, well, if they don’t install the ad blocker, then do we care.
Cory Underwood: But it’s a whole different scenario when the stuff starts doing that by default. It used to be that you know cookies would persist until either they expired somebody reinstalled the browser or someone bought a whole new device.
Cory Underwood: But now Firefox deletes no trackers every 24 hours so far, we might delete them after 24 hours.
Cory Underwood: Edge won’t even transmit a cookie if it exists if it’s two or domain that it thinks is trying to track you and you don’t have sufficient slight engagement score.
Cory Underwood: So,
Cory Underwood: Different things could potentially still work if they’re built differently.
Cory Underwood: But at the same time. That’s going to force companies to reinvest a huge amount of money reworking what their website architecture looks like.
Tim Wilson: I have another another question about the server side tracking. Does that, does that solve everything.
Cory Underwood: Which then sends that out to additional targeting, but that the fact that the clients involved at all means it can potentially break so it gets further because it’s not calling those endpoints directly, but at the same time, it’s not as
Cory Underwood: Probably long term viable, as you would be if you did server to server communication because then the browser doesn’t see it at all.
Tim Wilson: So I’m fast and furious. I could I could read them, or if you want to start cherry picking them. They’re good.
Cory Underwood: Next one is break site usage features like remembering your login or return to your bank site so
Cory Underwood: And so limits the destructive potential of that technique by artificially capping how long that persistence can stick around.
Cory Underwood: They do state. However, that if you’re going through the effort to retool your server to do that. Specifically, they’ll let that through so
Cory Underwood: From a security purpose so
Cory Underwood: They won’t break that yet.
Cory Underwood: However, I think that ultimately we could see a large scale shift in the next couple of years if they figure out how to do authentication without relying on cookies.
Cory Underwood: I don’t know what dub key is so I don’t have a point of view on it.
Tim Wilson: Jump in, Tony.
Tim Wilson: If you want to
Tim Wilson: unmute
If you can
TonyZ: Yeah. So my understanding, thanks for thanks for talking about this stuff. This is a new proposal from Google about
TonyZ: A new version of the privacy sandbox that appears to be a bit of a power grab from Google. And so I haven’t looked into a very deeply. But it was called Sparrow and now it’s called dev key it’s using these various Turk Soviet era code names for all of their Skunk Works. Thanks, but that’s
TonyZ: It’s apparently the new privacy sandbox initiative.
Cory Underwood: So their privacy sandbox and interesting because it actually
Cory Underwood: Allows them billions of potential identifiers, as opposed to the number of bits and identifier, so
Cory Underwood: Maybe now the thing is if Chrome it opposite, but nobody else does.
Cory Underwood: Then
Cory Underwood: You have a solution that works for one browser. So it doesn’t really yet you what you’re looking for, because, well, they control 65% of the most of the market entirely in North America Safari like iOS is actually 53% of the mobile market.
Cory Underwood: So as more and more stuff transition to mobile.
Cory Underwood: The potential fix that offers
Cory Underwood: Will likely exceed 50% unless they can get Safari to play ball and if they go too far in one direction, Apple doesn’t agree that we’re going to see a divergence and it could be that they just can’t.
Cory Underwood: Market to Apple devices years
TonyZ: Make sense
Cory Underwood: Which is something that
Cory Underwood: Facebook’s dealing with because Facebook said that when the app changes going next year. They’re gonna have to see.
Cory Underwood: What they’re skeptical. It would even make sense to offer their audience network on iOS, because they don’t think they’re going to get enough of a match rate to be able to do anything.
TonyZ: So there’s got to be some kind of publisher revolt. No.
Cory Underwood: Yeah, well, I mean that’s that’s possible, but it’s not like Apple needs the money so
TonyZ: It’s a good point. Thank you.
Grace Koplow: Kind of goes to the next.
Grace Koplow: Question I don’t. I would agree that it’s not just a technical problem we are having lots of conversations you’re even
Grace Koplow: Anywhere from technical solutions to how do we even just change how we optimize and do we need to rethink our optimization strategies to
Grace Koplow: Accommodate this and it might not even be again a fix it might have to be a reframing of how we collect data or interpret data or what data we’re using to make those decisions.
Grace Koplow: Fun times
Cory Underwood: So I agree with
Cory Underwood: Phil, it’s, it’s not just the technical problem. You got to really think about what your requirements are.
Cory Underwood: And
Cory Underwood: The important thing to realize is your dev team didn’t build it incorrectly to begin with a lot of this stuff has been deployed over like the last year to two years. And so what likely was seen as a viable long term strategy three or four years ago just may not be the case.
Cory Underwood: And so as the scenario changes. The company has to be willing to like write off what they previously done and possibly re implement things that they had working just to keep the site working what you did.
Cory Underwood: And then I agree that
Cory Underwood: Things are likely going to be rough for
Cory Underwood: Free marketing remarketing and programmatic
Cory Underwood: Programmatic is being reviewed for legal eligibility in the EU, so we’ll see if that falls the barrel of GDPR
Cory Underwood: But
Cory Underwood: A lot of this stuff comes from knowing something about the customer. So you can tailor the big like you know that they had previously seen something so you can spend more.
Cory Underwood: Or they’re in certain areas so certain classifications of programmatic might stop being viable. So you can’t do bid adjustments off certain variables, just because you no longer have access to them.
Cory Underwood: So if a site requires login to enter. How will ads optimization in the pack analytics be impacted by these changes. It depends on how those respective platforms work.
Cory Underwood: So,
Cory Underwood: On Demand generation of the page, then that’s probably going to work a lot better at least under the current architecture.
Cory Underwood: And we’ll see how to see. So like for example conductors to concise expect offer server side testing solutions and because of the way the tests are built. They don’t care about this.
Cory Underwood: You have some server side, like if you happen to take Google Analytics and you fired all of your measurement hits through measurement protocol, rather than through a Tag Manager, you might not even notice this.
Tim Wilson: We have a pause. So I have while this was going on. Definitely confirmed that I left up and did not get breakout rooms re enabled on my
Tim Wilson: Account. So, we will not be doing the breakout room portion of this so
Tim Wilson: I now do have a to do on my list to get that figured out in the next week. So if you come back in November.
Tim Wilson: I will, I will. I got two strikes, but I’m confident that I will be doing that correctly next month.
Tim Wilson: So what I will, we’re going to do is pause and do the door prizes.
Tim Wilson: And then we’ll kind of just hang out with more questions. So if you have questions and thought you didn’t want to take the time to type your question because he didn’t want to miss what was being said
Tim Wilson: I’m not going to drone on for a few minutes. And now’s a good time to type of question and then we’ll hang out for a few more minutes.
Tim Wilson: And I am not going to try to
Tim Wilson: Pull up another slide because all the slide says is that our door prize is $50 doordash gift cards, since many of us are most of us are still not out and about and that’s the closest we can get to providing food to a few people.
Tim Wilson: So we work with Price Waterhouse Coopers for a fully audited process for random selection of winners.
Tim Wilson: It’s maybe not quite that rigorous, but we do actually have a process we feel pretty good about for randomly picking our three winners and the winners of our $50 gift cards must be present to win.
Tim Wilson: Well, Jason. Looks like we lost one
Tim Wilson: I think Brian Foley is gone.
Tim Wilson: Okay, so we will move
Tim Wilson: On down the list. The next winner is.
Tim Wilson: Dustin Bowden who is still here. Outstanding.
Tim Wilson: And the next one is Josh borstein also still here. Congratulations.
Tim Wilson: And then the final winner is Kaylee night, who is also still here. So congratulations. I will be private messaging you to get the email address to send your doordash gift card to you’ll have a
Tim Wilson: Moderate meal on on web analytics, Wednesday, at some point, and you can think fondly and maybe you should order cookies with whatever wherever you order from it should be somewhere that has cookies to make up for this topic and the depressing depression of it.
Tim Wilson: So with that we’ll go back to look at that back to the FTP timeline, so
Tim Wilson: I will turn it back over to you guys for some
Tim Wilson: More questions will probably go for as long as the questions are flowing, but probably no more than 10 or 15 minutes and go from there.
Tim Wilson: And I’m going to keep recording this, because they seem like they’re good questions, leading the discussion.
Cory Underwood: So I do see Apple working with data protection authorities, because they will likely be required by law to do so.
Cory Underwood: And if they want to continue operating in that country.
Cory Underwood: They probably have to play nice with the government.
Cory Underwood: So that’s where I think that’s going to end up
Cory Underwood: Then Tim correctly said it’s dangerous to think of finding workarounds because it’s not necessarily a work around as much of it is a shift in architecture that is not subject to those challenges. So the like the types of constraints are changing.
Cory Underwood: With that said, there are other valid architectures that don’t get subject to those constraints. So it’s kind of like
Cory Underwood: an igloo is a viable piece of construction, if you are in a place full of snow, but it is probably not going to work for you in the desert.
Cory Underwood: Then ducky apparently is instead of reaching out to add servers, the browser does all the calculation, and I would be super interested in how that
Cory Underwood: pans out be cuz I would think that if you have everything being done on the client. It is
Cory Underwood: A security nightmare. When you’re dealing about things with money and how much to charge people for showing an ad
Cory Underwood: Because if I can modify the client then potentially I could drive up what that looks like. So
Cory Underwood: Be be interested to see if people break that in how quickly that happens because ad fraud is a thing.
Cory Underwood: Does this affect web analytics versus add tags in the answer is, it affects both potentially in different ways. So for web analytics, depending on how its constructed
Cory Underwood: Your primary reporting suite can skew certain pieces of data, particularly when you’re doing things like cohort analysis or cross session analysis things that require
Cory Underwood: Some concept of history, like you’re building a timeline. If you never have that starting point. You only know what’s the present. So, things get bucket incorrectly and stuff like that for the add tags. They might also have that issue. But then they also just might not work so
Cory Underwood: For example, Firefox aggressively blocks known trackers from getting cookies and if that’s how the particular ad tech works. It needs a cookie, even if it’s first party. It doesn’t get access to it, it doesn’t get transmitted, would that network request.
Cory Underwood: It does not work. So even though you’re calling the ad service the ad servers doesn’t have that initial identifier to say, hey, this thing happened from this person.
Cory Underwood: And in that instance, if one is working right, and the other is not. You could see a pretty sizable delta between the two.
Cory Underwood: Conversely, if you had one working right like your analytics platform and the ad tag wasn’t working, you would think.
Cory Underwood: Not only what you see a delta, but the analytics platform could potentially get the attribution right even though, if you look on the ROI calculation in the vendor platform. It’s going to be probably potentially some order of magnitude less, depending on your exact browser mix break down.
Cory Underwood: tags for third party vendors would be affected.
Cory Underwood: Depending on how they work. And if they are classified as tracking
Cory Underwood: So Firefox uses like a list done by disconnect me to identify what it knows as trackers, and then out of that it goes through some calculations to figure out how it should
Cory Underwood: How it showed or if it should even mess with that system edge does something similar with their trust protection lists. So depending on how they work. They might be impacted.
Cory Underwood: It is time to start industry standards for analytics to work with browser vendors and there is a GitHub group that you can join to discuss all of these privacy issues with the people who are designing the solutions for how to solve them.
Cory Underwood: And I can get that
Tim Wilson: Good Hope you said GitHub group.
Cory Underwood: There is it. Get a group
Cory Underwood: Oh, and I am subscribed to the repo and so I get emails. Whenever someone comments on something.
Tim Wilson: Just for those who don’t know Till’s attending from Germany. So, middle of the night.
Tim Wilson: For him.
Cory Underwood: I do think that the pending antitrust investigation may result in changes for Google on how it collects and or shares data if
Cory Underwood: Congress gets the FTC to engage the DOJ to bring suit.
Cory Underwood: Whether that happens or not as a big question mark. And so the other possibility would be bipartisan
Cory Underwood: Bill passing in Congress for a data privacy law.
Cory Underwood: which so far has failed multiple times. So I think, given the current administration and the legislation.
Cory Underwood: The chances of that happening are slim, the elections could potentially change the control of the Senate and the Presidency, so depending on which ways that go, I would say that is either more or less likely in the next four years.
Cory Underwood: Ritual is always cool
Cory Underwood: And I don’t think you need to go all the way back to reading log files, although if you have that skill that could ultimately come before fall
Tim Wilson: We will will make sure that Korean track down the GitHub group and we will post it on the will do like a show a show recap, or will have the recording of this video and we’ll
Tim Wilson: Make sure we’ve kind of track down though, right, that’s something that could be. That’s a group that the link can be published. Right. Yeah.
Cory Underwood: Yeah, anyone there and they want people to go and chat with them.
Cory Underwood: So if you don’t, if you have a strong viewpoint on something that they’re like they’re thinking about doing and you don’t voice that
Cory Underwood: And then they do it.
Cory Underwood: Well,
Cory Underwood: You might have been the voice that caused them to differently. And if you got them. That’s probably your fault.
Cory Underwood: So I encourage you to engage with the group.
Cory Underwood: There is a complaint that Google is unilaterally driving standards. However, there is also a complaint that Apple’s unilaterally driving standards so
Cory Underwood: Have doesn’t have one, you know, six to the other.
Cory Underwood: What are the chances that any antitrust settlement will be intelligently thought out and I would go with close to zero.
Cory Underwood: The thing is though it looks so much of this
Cory Underwood: Gets into the weeds very quickly and
Cory Underwood: I am skeptical of people without a technical background getting enough of the context to think it through in a way that would make sense.
Tim Wilson: Well, and this is fair. I mean, this is we sort of had it with the two of you kind of really kind of as a as a way to speak to this as you’re talking to clients who are grappling with this.
Tim Wilson: You are kind of more the bearer of bad news. And it’s bad news about if we do nothing. Things get worse, right like that, that can’t be a message that anybody want congratulations.
Tim Wilson: We do exactly what we’ve been doing will be delivering less value, you need to spend money invest time and resources to not even quite even maintain the status quo.
Grace Koplow: Get kinda
Grace Koplow: But it’s it’s it’s a it’s this big push of people apparently care if they’re being followed by online data and Facebook and all this sort of stuff. And it’s raised quite a bit of
Grace Koplow: Chatter from everybody to try and prevent that. And so with power, comes great responsibility. So as these things were allowing us to track it. And those things. It’s now wiring. I think people are
Grace Koplow: Reacting and pulling some strings to make it tighter. So your point. Yes, if we do nothing. Things are going to break in your data is going to be wrong, and the things that we’re doing. It’s not a golden ticket, but at least it’s better than doing nothing.
Cory Underwood: So I just posted the link to the GitHub group in the chat, because I found the wilderness was talking if 24 passes.
Cory Underwood: It’s not illegal to capture data points under the current proposition of the law. If you want to read all 53 PAGES OF IT. IT IS ON THE BALLOT pedia link I posted earlier.
Cory Underwood: But with that said, customers would get more control over what they can and what you can and cannot use there.
Cory Underwood: So,
Cory Underwood: Regardless of the technical design that your site is subject to you would need to comply with the regulation for
Cory Underwood: Servicing people in California. If you are one of the groups that that law applies to
Cory Underwood: And as far as for how soon. This rolls out California typically leads the way in the United States because it has one of the largest market share, like
Cory Underwood: They drive a huge segment of the economy. Right. So, then your company gets have the fun conversation. Do we standardize to the most restrictive data. Data laws that we have to deal with, or do we make all of the processes insanely more complicated by treating California is special.
Cory Underwood: And there’s some pros and cons to each
Cory Underwood: Now obviously if you’re going to shoot California as special then the training and stuff goes way up because now you’re dealing with multiple scenarios.
Tim Wilson: I think I remember Stefan ML trying to take just on the from the regular regulatory side trying to line up and he think he concluded that would be impossible to behave in a way that complied with all current regulations.
Cory Underwood: It’s not because
Cory Underwood: For example, CPA is opt out GDPR is opt in. So if you are trying to align to the CPA, you are against the GDPR but if you’re complaining to the GDPR then you are losing data that you could otherwise having with the CPA.
Tim Wilson: But I think he was going a little deeper than then kind of just the opt out opt in and a little bit more even kind of the data handling.
Cory Underwood: Right, yeah.
Cory Underwood: And Japan just updated their data privacy laws this past June and Brazil has a huge data privacy.
Cory Underwood: Initiative that they published. And so if you are doing business internationally, particularly if you’re transferring data out of Europe, you should be very concerned and you probably want to look at the most recent court rulings.
Tim Wilson: Ah, well,
Tim Wilson: I can’t
Tim Wilson: Wait, what
Cory Underwood: I would have to look
Tim Wilson: Perfect. Okay.
Tim Wilson: Well, I think.
Tim Wilson: This is probably a good time to wrap. I kind of neglected to actually thank both of you for doing this when you were done with the main presentation because we slid so smoothly into
Tim Wilson: QA, but
Tim Wilson: Thank you both. Grace and Corey for
Tim Wilson: Presenting and the discussion and thanks to everyone who attended with, I don’t know that we’ve had this many string of of solid and engaged questions at a
Tim Wilson: webinar next Wednesday. So
Tim Wilson: Hopefully you found it useful and fun and three of you will get food out of it and
Tim Wilson: Hopefully we will see you next month if you register for this through the website or have registered in the past, you will get emails.
Tim Wilson: A couple of weeks out from our next event.
Tim Wilson: If not, you can sign up on CBS W w.com and you’ll get notified of future events, but hopefully someday we will be back in person, although it is kind of cool to have people from
Tim Wilson: All over the country in the world attending. So that is kind of the one silver lining to this is in and actually say Grace’s in New York and Corey is in the middle of nowhere in Michigan.
Tim Wilson: That fair
Cory Underwood: Actually in a little more
Tim Wilson: But so thanks everyone for coming and hopefully we will see many, if not most of you next month.
Cory Underwood: And I would say it’s not snowing here yet. But I would say the biggest thing that I would say in addition to the browser privacy stuff we’ve discussed is
Cory Underwood: Apple’s changing the app guidelines for data collection in iOS next year and they will become very close to GDPR. So even if you are in the United States. If you are doing any sort of data collection through iOS. That is a process that you will need to undertake.
Grace Koplow: I like that you just drop that bomb you’re about to leave.
Tim Wilson: Thanks, everyone. Oh, yeah.
Grace Koplow: Oh yeah, all your fire really respect.
Tim Wilson: You tell my mouse was hovering over the end button.
Tim Wilson: All right, thanks everyone. Have a great rest of your evening and rest of your week. Bye.
Lincoln kicked off with a definition of local SEO:
Local SEO is optimizing your website and online presence to ensure visibility when users conduct a search with geographic intent.
That “geographic intent” piece has moved well beyond the “searches that include ‘…near me.'” Because, well, Google’s gotten smarter than that. For instance, a search for “lawn mower repair” is likely to have geographic intent. Depending on the industry/sector, this sort of thing can happen all the time!
At the same time, local SEO isn’t really a singular thing that an organization should focus on:
It’s part of an overall SEO strategy (including technical SEO)
It has content marketing implications (of course!)
It may lead you to focusing on things like user reviews…which have a halo effect of benefits beyond just SEO
The Google algorithm is really trying to triangulate on a few different things, all of which can be assisted with different SEO tactics:
Proximity—is the business near the user’s location?
Relevance—does the business offer what the user is searching for?
Prominence—is the business known (inbound links from quality sources) for what the user is looking for?
There are lots of tactics that play into this, including establishing and maintaining Google My Business (GMB) for the company and its locations, soliciting reviews (with some specific focus), link building (with the right sites), and, of course, the site structure and content (such as dedicated pages for each location for multi-location businesses).
No good presentation would be complete without recommending some resources, and this was a great presentation!
We recorded the event (almost all of it; we whiffed on pressing the record button for the first few minutes!):
This month’s event was over a year in the making, but it was well worth the wait! It was a joint meetup of R-Ladies Columbus and Columbus WAW:
For the R crowd, it was an introduction to digital analytics data
For the digital analytics crowd, it was an introduction to R
For the crowd that had already been crossing over, it was an evening where the peanut butter and chocolate finally came together to make an Reeses’s cup!
Dr. Katie Schafer from Covail provided an introduction to R using a (still available!) RStudio Cloud project and then walked through a set of Google Analytics data all the way from some basic exploration of the data to a linear model attempting to predict the drivers of revenue per visit (RPV). The model itself was not very “good,” but the point of the exercise was to provide a basic introduction to the platform and approach, as well as to show how to both run and evaluate a model.
Some of the basic exploration of the data…and of ggplot and faceted plots:
Some more exploration using a scatterplot:
And some of the results of a generalized linear model:
It was our third virtual event, and we’re slowly managing to find our groove in this meetup-by-Zoom era. With an unexpected boost by way of a tweet from @googleanalytics, we had a strong and engaged showing, including attendees from four continents (North America, Europe, Australia, and Africa)! The crowd was not disappointed, as Mai AlOwaish from InfoTrust provided a concise, but quite thorough, rundown of the ins and outs of Google App+Web.
A smattering of takeaways:
As a platform that draws heavily on Firebase Analytics, it is a fundamentally user-centric platform that relies on an event-based model. Fare thee well “all hits are of type: pageview, event, or transaction!”
The event-based model has great power…but also will take some getting used to for those who are steeped in a long tradition of Universal Analytics
The platform has been in beta since July 2019. But, what does “beta” even mean in Google terms? Don’t they keep things in beta when they’re clearly ready for production use? Well, yes…but it would be pretty risky to just deploy App+Web at this point. Functionality to match, for instance Enhance Ecommerce has not yet rolled out (but there are many signals that it is imminent)
Unlike the cutover from the original Google Analytics to Universal Analytics, dual-tagging a site with both Universal Analytics and App+Web is not a problem: they use different measurement protocols and send data to different realms within the Googleverse.
App+Web—a free platform—has native connectivity to Google BigQuery (one of those things that used to stand out as a differentiator between free Google Analytics and GA360). Of course, depending on the volume of traffic you’re sending to BigQuery, you may wind up with a bill for that, but the analytics platform itself does not incur a cost.
The “Analytics Hub” adds a more robust analyst tool for working in a browser to drag and drop dimensions and metrics and do handy visualizations like Venn diagrams to visualized the overlap between two segments.
And much, much more! But, you can just watch the video of her presentation for a full run-down (or check out the slides):
There were a few notable links shared in the chat for more information:
This meetup was, once again, also a fundraiser for the Mid-Ohio Food Collective. With help from our sponsors we’re matching the first $1,000 donated! Thanks so much to those who have already donated, it means a lot to us and even more to the people you’re helping. We’ve still got matching funds available, you can donate here now. We’ll be running the campaign for the rest of the month so it’s not too late!
Thanks to everyone who joined us for our May meetup! While virtual events certainly have their downsides, one of the nice things is being able to bring in speakers (and attendees) from all over the place! This month we welcomed Michael Helbling from AJL Analytics in Atlanta to speak about emotional intelligence. Michael is also the co-host of The Digital Analytics Power Hour along with Moe Kiss and our own Tim Wilson.
This meetup was also a fundraiser for the Mid-Ohio Food Collective. With help from our sponsors we’re matching the first $1,000 donated! Thanks so much to those who have already donated, it means a lot to us and even more to the people you’re helping. We’ve still got matching funds available, you can donate here now. We’ll be running the campaign for the rest of the month so it’s not too late! Don’t make Tim have to start breaking into your morning news coverage with tote bags and guilt!
Analysts absolutely need this kind of intelligence in order to have our work truly listened to, and yet like our reports themselves emotional intelligence is frequently ignored or looked down upon. Perhaps it’s because analytical types don’t tend to be as naturally strong in this area, or perhaps it’s because in general this sort of intelligence is typically labelled in a pejorative way as “soft skills”. In any case, this is an important discussion to have!
Another one of the advantages of virtual is that we’ve got a recording of the session, so who cares what I thought of the session, watch it for yourself right here!
Maybe it’s Michael’s high degrees of self-awareness and empathy, or maybe it’s the fact that he is from Ohio originally — but Michael’s featured example of someone who displays strong emotional intelligence is Ohio’s own Dr. Amy Acton. Let’s face it, our jobs are a heckuva lot easier than hers, and if our stakeholders ignore the data we present people’s lives aren’t on the line — so we can learn a lot from her presentation of analysis! Michael also highly recommended this piece on Dr. Acton from the New York Times. Michael’s “Dr. Amy Acton Fan Club” lawn sign is in the mail.
As usual, we had some great questions as well, some of the things that came up in the QA were:
In some ways, April was just like any other cbuswaw meetup: interesting content, 40-50 digital analytics folks, great questions and community. But pretty different in other ways, as this was our first ever virtual event!
Thanks to our speakers Monica and Anthony Weiler from Stratos Innovation Group as well as all of our participants for joining us in this experiment. We had a good time and hope other attendees did as well, please reach out to us if you have any feedback.
We had many regulars show up as well as people from all over: Florida, California, and even a strange far away place where everything is so different yet somehow familiar… Michigan. Oh, and someone from Australia was there too.
Most importantly, we’re all here to learn about interesting topics, and Monica and Anthony delivered on that. They explained both what customer journey mapping is as well as how the principles of service design fit into this kind of practice. Ok, and they showed us a lot of really cool looking maps.
Seeing the beautiful and detailed customer journey maps that Stratos creates is impressive, but how the heck do you get from some basic usage data and knowledge about your customers (that probably sits in a bunch of different silos) to something like that?
Monica and Anthony walked us through their approach, bringing in lots of different disciplines including: psychology, human factors design, and traditional behavior analytics. Their holistic approach blends hard data along with many harder to directly quantify factors… an art form not unlike many forms of data storytelling.
Do you need a PhD in Integrated Systems Engineering specializing in Human-centered Design, Customer Experience and Organizational Change Management (yes, that’s what Monica’s degree is in) to do this work? Well, we’re sure it doesn’t hurt, but no! Monica and Anthony have helpfully provided some additional resources to start getting your feet wet in the form of a free ebook on their site.
They have also provided their slides! Many of the maps are of course under NDA and left out of this version of the deck, but they have generously offered a walkthrough of these projects if anyone is interested.
Please join us next month, when we expect to be virtually meeting up again in similar fashion — topic and speaker TBA.
If you’d like to speak at an upcoming event (virtual or otherwise) or you have suggestions for a speaker or topic, please fill out our speaker call form here!
Our February meetup was a fun and informative session on factor analysis with Ahmad Ahmad downtown at Hopewell. Thanks to everyone that showed up even though the weather was kind of awful! Ahmad first gave us an introduction into the concept of factor analysis and when it might be helpful, then in good cbuswaw style proceeded to show us some real data and analysis, throughout fielding some solid questions from the crowd.
Some data is easily reducible to a smaller set of groups. For example, during Ahmad’s talk there was a significant amount of water droplets falling from the sky. These droplets could be very easily reduced into different types: rain, snow, freezing rain, and sleet. This is a case where we definitely don’t need factor analysis, because these are directly observed variables, i.e. we know what precipitation is what without using any statistical methods. Nobody would ever ask an analyst to quantify if it was snowing or not, but they would ask us to figure out from a bunch of different web stats why their users aren’t converting. How do we boil those dozens of dimensions down to groups of a few useful ones that share some kind of common underlying dimension that was not directly observed? And then what happens with even less obviously reducible data? We need statistical tools! Ahmad first walked us through a couple of thought exercises on this kind of dimension reduction to see how this might function from a high level.
Ahmad knew to bring data though, so he brought an analysis that he did based upon Boeing employee survey data where he attempts to turn 30-something different questions into a useful number of higher level groups, in this case 4. He did the analysis in SPSS, but he brought the results into Excel for presentation and has been kind enough to provide that Excel file as well as his slides:
Those of us that also saw Dr. Michael Levin talk on cluster analysis last September may also want to revisit his talk to remind ourselves of the differences between these two very similar topics that go hand-in-hand but have different uses. Factor analysis is about discovering these underlying groups (like finding the survey questions that could be represented into an underlying factor of “career stage” — age, expected retirement date, and less obvious dimensions like perception of selling potential). Cluster analysis is about sorting the surveyed people into groups by using the existing dimensions, such that group members are more self-similar than people in another group.
Please join us next month when we’ll be back at Rev1 learning about service design with Monica and Anthony Weiler.
We kicked off the new year at Otterbein University, we’ll be at Hopewell next month and then back our regular digs at Rev1. In a well-attended lecture hall Elizabeth Eckels, CEO & Founder of Bancroft Digital, presented an intro class into what the world of consulting is all about. Not about the tedious details of how to setup an LLC or your corporate tax structure, but a lively conversation about analytics consulting practice itself, with real talk about the pluses and minuses.
The audience was fully engaged for this one, as basically everyone has either spent some time as a consultant or engaged with a consultant as a part of their job. There’s a ton to learn from both sides; learning how to most productively engage with your consultant is important too! Going out on your own in whatever the form can present a lot of risks and benefits, but Elizabeth walked us through both in a thoughtful way.
A few of her tips:
Don’t undervalue yourself.
Especially if you work at home, setting work boundaries can help keep you sane.
Frequent and thoughtful client communication is incredibly important. Put yourself in their shoes, but keep records to protect yourself too!
Nobody knows if you’re wearing pajama bottoms on a video conference, unless you have to stand up.
She has also kindly provided her slides! Check out slide #18 for a list of resources.
Welcome to the many new faces this time, hope to see you at future events as well! Also welcome back to one popular canine face, Elizabeth’s dog Cash.
Please join us next month downtown at Hopewell where Ahmad Ahmad will present on factor analysis.