You Can Handle The Truth: Change Management Tips - David Kirschner, Independent Advisor

At Google, David witnessed first-hand the pros and cons of embarking on an attribution program. The pros included a greater focus on data hygiene, better planning and previously undocumented cross-department cooperation. On the downside, David saw how introducing an entirely new measurement strategy negatively impacted compensation plans, trust in the organization's data, and subsequently the ability to act upon it.

In his presentation, David will share advice designed to reduce points of friction arising from the introduction of new measurement initiatives. Topics will include how to communicate change, set goals, recruit talent and overcome objections.


David Kirschner

Independent Advisor, Former Google Lead

In 1996, David was asked if he knew how many people were visiting the Web site his company had just launched. That single question sent David's natural curiosity into overdrive. Over the next decade, he became one of the first enterprise-level digital consultants, working with top clients like Ford, Marriott and Anheuser-Busch.

After rewarding stints with, and Omniture/Adobe, David was tapped by Google to run their Attribution team, where he spent five years answering challenging marketing questions for the world's largest advertisers. In 2018, David became the VP of Data and Analytics for Zoosk, leading a digital transformation, and helping them achieve a successful merger. Today he advises clients remotely and is proud to serve as a Strategic Advisor to ObservePoint. He also remains involved in advocacy for the autism community.

David Kirschner

I'm eager to speak with you today about some tips for measurement change management. I've been doing a few things for ObservePoint, perhaps you saw my presentation about attribution a month or so ago. If not, please take a look at that, but just a little bit about me. I started in this business before it even was called web analytics. I've been doing this since the late nineties. I was one of the first hires at Omniture. I think I was their first ever remote hire. And I came to manage the Adobe consulting group from 2008 to 2013. From there I went on to Google where I ran their attribution team, both pre-sales and post-sales. And I was also briefly the VP of analytics and data for Zeus, the dating app. Right now I'm doing my own thing, Kirschner Solutions, which focuses largely on this type of change management project. So hopefully you have some questions at the end. I'll try to leave some time, but let's get into it.

First and foremost, who doesn't love change? Everybody. Nobody loves to change. This is not something that humans are really eager to do. We're kind of hard wired to do the same things, the same way. That's just how we are as human beings. So, change provides an opportunity to inject fear, whether the change is introduced at home or at work, it's always something that people need to adjust to. And it's very rarely welcome even when the goals are something that is, something that everybody agrees that needs to change. It doesn't always work that way. So before you start a change management project, think through everything.

I love this quote here at the bottom. "Never take down a fence until you know why it was put up." Coming in, you have to understand that the systems that are there are there for a reason, and you have to define your goal. There's a lot of different ways to approach this. I, kind of, sort of veered toward that golden circle where you start with the, why? The why usually for a measurement change management product or project rather is things like incompatible systems. Maybe there's been a merger or acquisition, certainly lately supply chains have been disrupted. Whatever the reason, it's something that you need to think through very, very carefully. And in terms of the process, the how, that's really where the rubber hits the road, you'll want to capture every idea. We'll talk about who's going to be in this group, that is putting forth these ideas, refine the ideas and vet the ideas. But most importantly, take stock of your current situation, document the you-know-what out of every system that you have, every piece of data that you collect, and really understanding how the processes work. From marketing data, you know, you really want to track the process that it follows from ingestion, right through activation and into optimization.

We used to joke when I was at that we'd have these planning to plan meetings, but you actually do need to have something like that before you embark on a project that's this massive, and potentially this disruptive. Timing and season now, they are important. This is not a great time to start a massive change management project. Not only are we in the middle of a pandemic, but we're also starting to come up on the holidays, PTO, yesterday was singles day, there was a huge retail holiday globally. So really the best time to kick off, I find, is Q1. You can't always do that, but it does avoid a few things. Particularly the dreaded New Year's hangover. I've been a part of a few change management initiatives that span two calendar years. And unless you identify a really clean break point a milestone where you can sort of stop and pick it up again, the following year and keep momentum, it's really difficult.

Everybody has a sluggish start and has stuff to do at the beginning of the year that can interfere with that. So as to ensure a clean cut-over, you want to make sure that your agreements have been reviewed. Any vendor changes they've been notified, all of your t's have been crossed and your I's dotted. You also want to make sure that you have a stress and roll-back test in place, because as they say, timing is everything. But then they say there's never a perfect time for anything, and that's absolutely true. So look for like a natural stop point where you could pick up a project again if it has to spend two years. While you're doing this initial planning as well, consider the ripple effect. Because measurement tweaks have massive impact. An example, I worked with a client who was doing a customer acquisition measurement and their measure was the signup plus 24 hours. What did they do in those next 24 hours? Did they buy, did they refer, et cetera? Problem was their concept of 24 hours was locked in their specific time zone, and this was a global business. So in some cases, people may have only had an hour or two hours or three hours to complete the action. And when we looked at the data through the "retrospectoscope," the last point on here, we found out it was off by a major factor. So make sure that you understand what the downstream impact is of the measurement changes.

Also compensation changes are important, specifically when I was talking about attribution. SEM managers aren't usually huge fans of attribution because it generally takes credit away from paid search and give some to display that can affect people's compensation, which can affect people's position, feeling about the company, et cetera.

Your performance KPI are going to be effected as well. Individuals may be taking time out of their daily tasks to learn this new system. Some people may be concerned that their place in the organization is going to be different than it was before, that their title or stature might be impacted. So these are all things that you need to consider before you start. And again, try to document this. About the "retrospectoscope" again, it's really dangerous because once you have a new system in place, sure, you can go back and look at all of the things that may have been erroneous, but there's not much value in doing that. The value is in going forward and making sure that the system that you have put in place meet your current needs.

So who is on the dream team? It starts with, and frankly it ends with, engaged C-Level Sponsorship. That's why that one's in bold red with the ugly asterisks around it. You really need a dynamic leader who has the gravitas and the weight within the organization to make this measurement change happen. They also have to be an enthusiastic cheerleader for the project, not just "rah, rah," but actually believe in the projects, goals, success - and oversee it. They should be the primary communicator to the organization, whether they are writing the emails or somebody is writing them for them. The C-Level Sponsorship is the key driver and the most important.

Also important are these department heads. These could be owners of a specific line of business. They could be a manager of a channel, et cetera, whatever, how the organization is broken out. These department heads have a greater responsibility to accurately and completely represent what their department's needs are that how this change will affect them. They're also going to be the ones who have to manage the performance and accountability through the change. And also driving adoption, surfacing issues, bringing them up to this team, as well as recognizing when individuals are really doing a great job with the product. But most importantly, they need to come prepared with that first bullet about knowing what their department needs, knowing how this is going to impact them.

The next group I always have involved in these teams are technical subject matter experts. These are the guys who ground the ideas that the other two have - the C level people, the department heads. We love to talk about the art of the possible, these people know what is actually possible, what is practicable. They can really, not only help get the program off to the right start and focusing on the right measures, but they can reinforce lasting change if you choose the right people, people who have stature in the organization because of their work because of their past accomplishments, you know, it really is important.

Now, it's also important to think ahead in terms of retention strategies for these people. If you're having some, somebody who you've identified as key indispensable, they're going to expect some type of reward for helping with this change management initiative. Whether it's a promotion, whether it's an acknowledgement, recognition, whatever it is, make that you have a retention strategy in place because not only are they and help you get there, these projects never really end. Change management is just a cut over to a new system and you'll need these technical experts.

And then finally project managers and auditors, just these people, you know, not everybody has PMPs in their organization, but some of the PMPs I've worked with have just made difficult projects so much easier because they're expert at moving things along, forecasting projecting, et cetera, you can call them auditors. If you don't have PMPs, people who are responsible for ensuring that the process is followed, that it's followed correctly and that any resistance is smoothed over. The nice thing about that is sometimes the project, or the technical SMEs and the department heads have a close relationship and you don't want to necessarily ruffle any feathers there. Sometimes the project managers or auditors can be brought into help have those difficult conversations with individuals on the team. But most importantly, what this does is it provides a standard measurement of the program success. These program managers will all be looking at the things through a similar lens and measuring progress similarly.

So now you're ready to create a project timeline. And as Willie Nelson said, the early bird gets the worm, but the second mouse gets the cheese. So analyze, figure out what you'd estimate this project to take, and then double DOUBLE THAT. Seriously. This is a kind of a joke in consulting, but when you're doing a change management project it's wise to do so for a few reasons. One, it enables you to create evenly spaced milestones. You don't want the first milestone to be in a week, and then the second one to be three months later. You want to be able to space them out and have them correlate to actual progress in the project. You also should expect timeline compression, either from that C-level executive or his cadre of other C-level executives. Nobody is ever satisfied with a projects timeline when it's first presented to them, and they want it done now. Be realistic about what it is if you have to give on that timeline compression, but don't give it away. Don't put yourself in jeopardy of not completing the project, just because of pressure. It's important for people to realize the severity of the change, the ripple effect of the change, et cetera. And of course there will be unforeseen delays, whether it's a pandemic or a supply chain issue or whatever it may be, there will be delays. You can pretty much count on that. So analyze, estimate, double is my advice there.

In my last presentation I talked about the One Source of Truth, which you need as well. When you're doing a measurement change, often you're changing systems, and there's usually a reason for that, but here's where you need to keep a single source of truth for each metric. It doesn't need to be the same system, but everybody has to understand why that system is the source of truth for that metric. And when you're looking at which of these should be the source, consider things like reliability, uptime, total cost of ownership, privacy.

And then this next one is probably the most important create and maintain a data dictionary. This is called different things at different companies, but basically to me, a data dictionary could be a spreadsheet or a Wiki, but something that details every stinking point of data that's collected: where it's collected, how it's collected, what systems it goes into, where it's stored, who's it accessible to all of these things. It's going to be a pretty long sheet, I think we're getting into the AH, AG type of spreadsheets here. But it's important to maintain it, because as there's turnover, as new people come on board and need to learn what a specific measure means and why it's used, this data dictionary is going to become critical. Also documenting processes end to end from the point of view of a user persona, is a really valuable way to kind of put yourself in the customer's shoes and make sure that the experience is amenable and what you would want it to be.

But when it comes down to it, it's all about relevance. People want to know, what's in it for me? Why am I doing this? What'd you want to say and what they're interested in, in the middle is that relevance. So when you're talking about relevance to people, you want to talk about how it's going to make their job better, how it's going to improve the company's stature, how it's going to make your department more important. Whatever the talking points are, you should think about them in advance, and they're not going to be the same for every single person. So you have to reinforce them. Yes, deadlines matter, but you're also going to acknowledge errors and be transparent. You're not celebrating failures, you're going to have to do that as part of a change management program. And the best programs are transparent about it, and admit when they've made a mistake and document it so that same mistake is not made again.

Celebrating intra-team cooperation. Here's where you can do things like quick wins, send out newsletters, awards, et cetera. Try to get the team to coalesce around the change and really mitigating preemptive change. That's a little bit different, mitigating preemptive changes. Taking a look at the group you have and determining, is this team capable of affecting this outcome? Do I need to bring in or augment staff? Do we need to pre-train people? You know, really trying to call out those things that could derail a change management project before it gets going.

One of those things is this guy. So anytime there's change, especially systems change, there is resistance. Law de la Résistance, and it could come from any number of people. Either the people built the system, they're the only ones that know how to use the system, they're the ones best at it so their job depends on it. And you're always going to have resistance for various reasons. Sometimes they just don't think the change management program that's being espoused is really valuable. So here's you have to be careful, you really want to listen and empathize while reinforcing that this is going to happen, and that they will have a future in the organization. Talk to them, it's a great time to have a career discussion, but invite these people to participate in the planning and especially the auditing. Because if they want things done right, and this is the right way to do it, and they want to participate, they can be a great help with them. But really helping these people prepare personally for the upcoming changes. This isn't going to be the majority of the organization. These will generally be the longer tenured employees, or again the people who are most familiar with the incumbent systems. But do expect, some resistance.

A good way to combat that is to over-communicate. You can't over-communicate during a measurement change management initiative. A few ideas that have worked for me, the dedicated C level newsletter that comes out regularly, like maybe every two weeks. It doesn't need to be written by the C-level executive, they're usually not the best writers, they usually write in bullet points which isn't ideal for everybody. But in that newsletter, you should do a recognition of milestones that have been reached and talk about the new benefits and what's coming up, who's in training, et cetera. Publish and discuss dashboards. One of the things that I found really effective when I brought a change management program into Zoosk, was we put television monitors around the office with certain KPI around the change management initiative. And what would happen was people would gather around the televisions while they were having their lunch and they would chat about it. And just fostering that discussion, keeping it top center, encouraging debate about the success of the program, these are all really valuable ideas that, while it might be difficult to do something like that at a co located workforce like a lot of us are today, you can still push stuff out in terms of visuals and communications. And like I said, you just absolutely cannot over-communicate, during a change management initiative.

The kickoff, now we're ready to go. We're going to generate excitement. A lot of things need to happen at the kickoff to make sure you get off to the right start. You need to impress upon the organization that this is the new way - lets celebrate this. This is the right way to go. Getting some type of commemorative event, some type of swag or something, but really presenting the vision with enthusiasm and concern. Recognizing that this is going to impact everybody in different ways. Setting up office hours, soliciting people on the team to volunteer to help others get up and running more quickly within the new system is also something that's been very effective.

Now, the other thing that considered measurement changes change a lot. And this is again talking about, not in the resistance, but maybe more of the ripple effect. People will have these questions. Do I have to learn something new? Don't maybe learn something new! How will this impact my compensation? Do I have a career here anymore? One size does not fit all here, this is why office hours are important. But in those discussions, you have to reinforce that compliance is key, rewards are coming, and make sure that they understand that they're gaining new skills by embarking on a new system. Maybe either bringing in Adobe, for instance, then they're switching over from Google analytics. That's a valuable skill to have that is portable to other jobs. So even if they feel like they're stuck here after this change management initiative can convince them that this is going to be valuable for them professionally.

And then there's Kanter's law, which basically says that in the middle, everything looks like a failure. And it's really small but the successor says "FAILURE. It may not build character, but it can destroy a spirit." And that's true. In-flight projects always look like they're off the rails. And while the beginning and the end are fun, the actual work in the middle isn't. So here's where again, that transparency and communication and honesty are the keys. Using those office hours to get a team pulse - how do you think it's going? What would you do differently? These are all valuable conversations that you really should have with the team.

Then we're starting to approach the finish line. An object in motion remains in motion, but getting over that final hump, those final few milestones are often the most challenging. So as you start to get wins, as you start to see the benefits of the change, use that moment to your advantage. Use use cases, acknowledge the people who are producing them. Use training and discovery periods as well to ask new questions, to bring new people into the fold. But really, as you start to see the first fruits of the project payoff, promote the you-know-what out of them. Because as Walter Elliot said, "Perseverance is not a long race; it is many short races one after the other." And that's kind of true. It's almost like a really, really long agile project. There are milestones, they're spaced all out further apart, but each of them is important and each of them can build off of the other.

Now when it's time to present the new truth, this is where the rubber really hits the road. Like these first presentations, where you're talking about the changes in measurement and here's the new truth versus the old truth is subject to a tremendous amount of scrutiny. So these first reports have to be accurate. They have to be. You can't prepare enough for these, cause any breakdown here in the metrics, systems, will cast the program into doubt. I've seen this happen many times. When I was at Google, we'd frequently do an attribution project, it would take months sometimes, and we'd come in to present our findings and say, "Here's what you thought the measurement was. Here's where it is." And we've had C-level executives literally slam their computer and walk out of the room because they weren't ready for that change. They didn't believe it, they weren't prepped. So if that happens, it's really difficult to regain confidence in the program. And anything after that is definitely going to be scrutinized. The worst case the program gets defunded or becomes the butt of jokes.

So the way to work around that is really to prepare your audience for this readout. Bring tons of backup data, whatever you think you're going to need, have it at the ready. Have the SME in the room with you to answer the questions that you may not be able to answer. Training resources, to talk about the challenges and progress in training. Don't get derailed when you're doing this, wait till the end for questions. And then just really review what's changed, and talk about what ripple effect that it has caused, but positive and negative. Most importantly, though, you really have to acknowledge that what this is is different than the prior reporting and everybody will need time to adjust. While there's always a hard cutoff date, there's always a period after that, that people are sort of getting used to the new system and figuring out what it means for them.

Now, we got this pandemic going on. Have you heard anything about this on the news, but it's kind of a big thing. So right now is really a difficult time. Shouldn't prevent people from stopping a change management initiative, but there is no business as usual today. YoY comps are pretty meaningless predictive technology, who knows? I mean, who knows where we're heading right now. Plus you've got work from home for everyone. My son is downstairs right now, I'm using his computer to do this and he's hopping on a zoom with my phone right now, which may be the challenge for everybody. So ambiguity is greater than it's ever been, but that should not stop you from doing a change management project if the time is right. Some of the tactics I suggested might not actually be feasible in a pandemic and you can't put up TV screens and everybody's house for instance, but you can do newsletters. You can do events. There's a lot of out of work musicians right now who would love to do a virtual event for your company, probably for pretty reduced cost. So get creative, you're going to have to because it is a new game.

In closing, change is no longer a choice. Covid-19 has changed everybody's method of doing business. And despite the challenges, there's a lot of opportunity. Portfolios are shifting into different directions, all kinds of new innovations are being brought to the fore, it's really not a bad time to be in business for some types of businesses, and they should take advantage of that. But the challenges are real, and they're not going to go away overnight. Final thought I'll leave you with is that people make things happen. Technology is an enabler, but as you're investing, invest wisely. People are, at the end of the day, what make the organization go and they're the heart and soul of your organization. So with that, I will stop and plug my site once more in case anybody is in a similar situation and, could use some advice. 

Previous Video
Build a Robust Data Strategy that Translates to Business Benefits - Tanu Javeri, IBM
Build a Robust Data Strategy that Translates to Business Benefits - Tanu Javeri, IBM

Learn what comprises good data and how to use that knowledge to build a customer strategy that works.

Next Video
6 Requirements of Holistic Attribution & Actionable Insights - Mikel Chertudi & Cameron Cowan, ObservePoint
6 Requirements of Holistic Attribution & Actionable Insights - Mikel Chertudi & Cameron Cowan, ObservePoint

Dive into the 6 requirements for generating actionable insights that drive effective consumer experiences.