Jeff Roussel built this session out of a pattern he hears constantly. As VP of Sales at KaiNexus, his days are spent in discovery conversations with customers and prospects -- listening to what people struggle with in continuous improvement. He hears the familiar challenges: employee engagement, visibility, culture building. But the single most common one, by his own account, is reporting. How do organizations figure out and measure what they are doing in continuous improvement? How do they know if they are being successful? How do they build a system that lets them see the results of their effort?
The session targets a specific, recognizable pain -- the year-end reporting scramble, where spreadsheets get stitched together, dashboards get rebuilt at the last minute, and leadership starts questioning whether the numbers can be believed at all. Roussel uses that pain as a way into the larger question: how to make continuous improvement reporting organized, accurate, scalable, and trusted -- all year, not just in December. He was candid that this session, arriving in mid-December, was too late to fix the current year. The promise was for the next one: that by this time next year, reporting would be less of a curse and the holiday season less stressful.
Mark Graban hosted, and introduced Roussel as someone who came to KaiNexus relatively new to continuous improvement and absorbed a great deal quickly -- a salesperson whose passion for improvement and for customers gives him an unusually sharp sense of what people are actually struggling with.
Jeff Roussel is Vice President of Sales at KaiNexus. He came to the company after nearly 20 years in software sales and -- as he described it in the session -- discovered when he arrived that he had been a lean thinker all along without knowing the term for it. Bottom-up organizational culture building was already his belief about how organizations should be run, and how he tries to build the KaiNexus sales team; continuous improvement simply gave him language for a belief system he could not have fully articulated before. At the time of the webinar he was approaching three years with KaiNexus. Through years of customer and prospect conversations, he has built deep insight into the most common struggles organizations face in sustaining continuous improvement, particularly around visibility, engagement, and reporting.
Mark Graban served as host and moderator. He is Vice President of Improvement and Innovation Services at KaiNexus, and an internationally recognized consultant, author, speaker, and podcaster, with more than two decades of experience in Lean management, continuous improvement, and leadership development across healthcare and other industries.
This was an internal-presenter session -- Roussel drawing on the hundreds of customer conversations a sales role generates to surface the reporting problems organizations run into and the practical ways through them.
Roussel framed the whole session around Simon Sinek's Golden Circle, from "Start With Why." Sinek's argument is that you have to move from the inside of the circle outward -- establish why something matters before explaining what you will do about it. So Roussel started with why continuous improvement reporting matters at all, and gave six reasons.
We tend to measure what is important to us. Many organizations pay lip service to continuous improvement and say they want to do it -- but the act of measuring it, and holding people accountable to it, is what signals that it genuinely matters.
You need to know where you are to know where you are going. Too many organizations fall into the habit of working as hard as they can and letting the chips fall, rather than establishing the current situation and building a plan to the next step.
Goals require measurement. The "M" in SMART goals stands for measurable. If you are going to set and achieve goals, you need a way to measure progress against them.
The "C" in PDCA means something. Roussel's point landed on a discipline gap: organizations teach their people to plan, do, check, and act in their work processes, but do not apply the same discipline to the improvement process itself. You have to slow down and check that the improvement work is moving in the right direction.
Reporting keeps executives committed. The improvement function is often jumping up and down trying to get attention for its work. When you can tie that work to business results -- show executives how improvement is moving the needle -- Roussel said the level of commitment, and the resources that follow, will surprise you.
Reporting helps in bull and bear markets. When times are good, being able to talk about your successes matters. But Roussel described a "bear market" too -- a leadership change, or an organization's commitment to improvement waning. Reporting, visibility, and justification are what let the improvement function communicate its importance and keep going through that.
Underneath all six is the instruction Roussel kept returning to: figure out your own why. Why is your organization doing continuous improvement, and why does measuring it matter for you specifically. Answer that before anything else, because the Golden Circle says you cannot answer how until you have answered why, and you cannot answer what until you have answered how.
The middle of the session worked through the "how" questions Roussel hears most often from customers, each paired with practical advice.
Nearly every customer wants to pull continuous improvement insights from across the whole organization, and nearly every one finds it either impossible or a Herculean effort to cobble together. Roussel named the root problem: people are trying to pull data from what he called a "Franken-system" -- a term borrowed from KaiNexus partner HubSpot, describing the mess of stitching together different solutions to solve what should be one cohesive problem.
In continuous improvement, that means a physical Kaizen board for daily improvements, an actual box for suggestions, emails about the boards and suggestions, photos of sticky notes, separate handling for A3s and value stream maps -- everything managed in different ways, then replicated across every department. Roussel's most vivid example: one customer with over 100 locations had each location email a spreadsheet every month, which a centralized process improvement group then had to cobble together just to count how many ideas had been added to the boards. As he put it -- he is not a lean expert, but that has to be one of the wastes of lean somewhere.
His advice, while explicitly declining to turn the webinar into a pitch: invest in some kind of system. Look at your IT group, look at platforms, look at KaiNexus. When organizations get the boards off the wall and into a digitized form, visibility into impact and results follows almost immediately. It is a two-fold problem -- first you have to get people to do the work in the system, and then the benefits of having a system around that work become significant. Doing it individually, across every team and location, is simply too much.
Roussel set the gold standard first -- everyone in the organization involved in improvement every day, the holy grail few have reached -- and the second principle alongside it: real improvement work should produce evident benefit to the organization. If the benefit is not evident, either the work is not great or the benefit is not being measured correctly.
His answer to trust is three metrics, treated as a set of checks and balances.
Activity -- what kind of improvement is happening. Activity tells you how hard the organization is trying, who really cares, and who may need encouragement to participate.
Engagement -- who is actually doing the improvement. Engagement is one of the best ways to keep score over time: tracking averages, comparing against them, seeing whether things are improving and whether the improvements are coming from one person or spread across the team.
Impact -- the actual results and benefit. Roussel called connecting improvement to business value one of the most requested things customers ask for. They track cost savings, revenue, and time, alongside more qualitative measures like safety, quality, satisfaction, health, and environment.
The reason to track all three together is that they cross-check each other. Engagement tells you whether all your activity is coming from one place. Activity tells you whether your impact was lucky -- one big improvement -- or the sustained result of a real improvement culture. Track all three and you get results you can trust, which is what lets you make better decisions. Focus on only one, and you create blind spots.
Roussel said this question is really two questions. The first: how are we performing -- are we getting better over time, is this year better than last, this month better than last month. The second: how should we be performing -- what does good look like, and how close are we to it.
For the first, his biggest advice was discipline around measuring often. Organizations have daily metrics, weekly check-ins, and monthly stoplight reporting for their work processes; improvement metrics deserve the same regular cadence. Knowing where you were lets you see what changed, which is what lets you make better decisions about how to move forward.
For the second -- measuring against the mean -- he was more cautious. Outside data does not arrive in a vacuum, and an organization at a different stage of its journey is not a clean comparison. But he offered KaiNexus benchmarks anyway, with a grain of salt. The implementation rate -- the percentage of all improvements where the organization actually makes a change -- had held steady between roughly 75 and 85% across customers for three years, and stood at 77% at the time of the webinar, ticking upward. If your number is far below that, it is worth examining. Across all KaiNexus customers, about 2.5% of ideas had a financial impact above $10,000, and about 1.4% above $100,000 -- which means you cannot simply ask for the big ideas; you need volume to produce them. Roughly, in 200 improvements you might expect four worth at least $10,000 or two worth $100,000, assuming you are implementing most of them. KaiNexus publishes an ebook detailing these and additional benchmark metrics.
Roussel's diagnosis here was that perfect becomes the enemy of good enough. People try to measure 15 metrics for every team and every department right out of the gate, lose track of which metrics drive success, and stall. His acronym was KISS -- keep it simple. Develop discipline around reporting before tackling the bigger issues. If that means starting with just a couple of measures, that is fine. A good starting trio: the number of improvements, their financial impact, and the percentage that go to a change -- track those over time and add mechanisms later.
And the best principle for getting started on improvement measurement is the same one you would hand anyone else: PDCA. Create a simple plan, test it over a couple of months, check not just how the improvement work is doing but how the reporting mechanism itself is working, adjust, and run it again.
The honest difficulty Roussel named: you care about the continuous improvement program, but not everyone cares as much as you do. His main advice was to think in terms of standards. A consistent approach to measurement is far easier to get people to follow. Everyone has a day job; rather than asking people to choose their own improvement metrics, you may need to provide the metrics they should track and focus their attention -- activity, engagement, and impact again. Making it easier for them, and limiting the mind share required, makes it more likely they will actually do the measurement.
He also recommended a dashboard system -- visual and consistent -- as a way to create a standard. When multiple locations all see a similar dashboard, and everyone knows others see it too, rolling the process out across a bigger group gets meaningfully easier.
Roussel pulled the "how" advice into three guiding principles for getting optimized around CI reporting.
Get organized. Not rocket science, he acknowledged -- but improvement reporting is so often scattered across slightly different spreadsheets in every department, pulled into a bigger spreadsheet, shared on a SharePoint site. With no real organization, getting executives to trust the data and getting people to participate is nearly impossible. It is a high-level principle, but it is a problem he sees again and again, and committing to it is one of the first steps.
Be accurate. If you do not trust the data, you cannot make good decisions. Inaccurate or inconsistent inputs -- whiteboard data that is not timely, spreadsheets that do not come in the same way every time -- mean starting from square one every time you generate a report. Even if you change nothing else, fixing the accuracy of your information leaves you better off.
Develop discipline. Roussel called this the hardest of the three and probably the most important -- not just for reporting but for improvement in general. Plenty of people call KaiNexus already knowing how to do reporting, and simply do not, because they have not built the discipline to do it regularly and hold people accountable to delivering metrics and keeping dashboards current. He recommended thinking in a cadence -- daily, weekly, monthly, quarterly -- where doing it daily makes the weekly easier, the weekly makes the monthly easier, the monthly makes the quarterly more impactful. He pointed to project impact boards reviewed in regularly scheduled meetings as the tactical form of that discipline: once people know the data is important, it becomes easier to hold them accountable to it.
Roussel closed with ten ways to improve continuous improvement reporting for the year ahead.
Start right now. Take a small team or one person, use a spreadsheet, and measure something -- and if you are already measuring, add a new metric. Like building a Kaizen culture, sometimes you just have to make the move.
Develop and teach a growth mindset. Roussel recommended Carol Dweck's "Mindset," and reframed the problem: it is not that you are bad at continuous improvement reporting, it is that you have not learned to do it properly yet. Teams that believe they can get better, do.
Focus on the three key areas. Activity, engagement, and impact -- know how much improvement you are doing, who is doing it, and what business impact it produces.
Measure against last year and against the mean. Know both how you are doing and how you should be doing -- a way to correct what needs correcting and celebrate what needs celebrating.
Set goals this year. Roussel said many organizations, asked about their continuous improvement goals, simply do not have them written down. They know they want to get better but have not decided what better looks like. Make them SMART goals, and set them now so next year does not repeat the drift.
Choose a few small wins to celebrate -- and do it this month or next. Roussel's favorite examples from customers: an award for the smallest financial impact improvement, which signaled to the organization that it is not all about money and that small ideas count; a "Wildfire" or "peanut butter" award for the idea most likely to spread to others. Get into the habit of celebrating wins, and people will reward you for it.
Develop a measurement standard. Technical and genuinely hard, but important: get people measuring activity, engagement, and impact in a similar way across the organization. It makes aggregation easier, and it lets people learn from one another because everyone is working within the same standard.
Develop dashboards. Stoplight reports, scoreboard dashboards, visual boards -- virtual or manual. Make sure people can see the information, and that they know you care.
Invest in a software solution designed to help. It does not have to be KaiNexus, Roussel said, though he would love it if it were. As improvement science matures, software plays a role -- breaking down barriers and helping people engage more frequently. If you are going to improve the way you do improvement, technology has to be part of the equation.
Listen to your people. The one Roussel said people forget most. The Kaizen principle is that the people who do the work know how to improve it -- and the people helping you measure improvement can tell you how to improve the measurement process. Ask them regularly, open the lines of communication, and they will pay you back many times over.
The Q&A drew out several useful threads.
Asked his own "why" for joining KaiNexus, Roussel gave the answer described above -- the discovery that continuous improvement matched a belief system about bottom-up culture building that he could not previously have articulated.
On benchmarking, an audience member asked whether comparing yourself to others is limiting versus comparing yourself to your own best. Roussel agreed it can be dangerous to compare only against outside entities -- the data does not come in a vacuum, organizations track different things, and one company implementing 90%-plus of its improvements while another implements far fewer does not mean the second is doing bad improvement work. Outside data can help you know whether you are playing the game correctly, but measuring against yourself is the more effective path to improvement over the long run.
Graban added a point about who you benchmark. Organizations often try to benchmark against companies exactly like themselves -- his joke was the small Vermont dairy producer searching online for another small Vermont dairy producer with a left-handed CEO. You can stretch your horizons by benchmarking against organizations that are not like you. A health system benchmarking ideas implemented per person per year against other health systems might aim to go from two to four; benchmarking against Toyota or an automotive supplier, it might realize the bar could be 50 -- comparing itself against an ideal rather than against a near-twin.
Graban also raised goal-setting, previewing his January webinar. Organizations often set goals suspiciously close to last year's -- his example, a patient satisfaction goal of 3.52 out of 5 when last year's was 3.49. Taking variation and noise in the data into account, 3.52 and 3.49 are essentially the same number; 3.52 is not meaningfully better. Set goals that are genuinely challenging, and use them to spur more improvement.
Roussel added a finding that surprises him every time KaiNexus looks at the data: industry does not seem to matter. There are top, middle, and bottom performers across every industry, and you cannot meaningfully benchmark performance based on industry alone.
Asked how he applies activity, engagement, and impact to his own sales organization, Roussel gave a candid answer. Sales has clear activity metrics -- calls, emails, meetings, demos -- which he tracks aggressively as leading indicators. Engagement is harder to make scientific on the sales side, so he handles it qualitatively: he "goes to the gemba," sitting in on sales calls, listening, and giving feedback. Impact is easier, because revenue is directly measurable. He also noted the sales team eats its own dog food -- meeting daily to discuss improvement opportunities and tracking them in KaiNexus -- while admitting they were not yet good at quantifying the impact of those changes, which was one of their own goals for the year ahead. Graban's close on the point: lean is not just for operations; lean thinking can be applied to everything.
This session is, by Jeff Roussel's own framing, about a problem first and a tool second. The reporting curse is real whether or not an organization ever buys software, and Roussel was explicit that the webinar was not a pitch. But the structure of the problem he described -- and the three principles he landed on -- map directly onto what improvement infrastructure exists to do.
The Franken-system is the clearest connection. A physical Kaizen board, a suggestion box, emails about both, photos of sticky notes, separate handling for A3s and value stream maps, replicated across every department, then aggregated by hand -- that is the status quo Roussel was describing, and the customer with 100 locations emailing monthly spreadsheets is its logical endpoint. The reason year-end reporting becomes a curse is that the data was never in one place to begin with, so December becomes a frantic exercise in cobbling. A purpose-built platform addresses that not by adding a feature but by removing the cobbling: when the improvement work is captured in one system as it happens, the year-end report is a query rather than a reconstruction. Roussel's own advice -- get the boards off the wall and into a digitized form, and visibility into impact follows almost immediately -- is a description of what that infrastructure does.
His three trust metrics are a platform's reason for being. Activity, engagement, and impact only function as checks and balances if all three are captured consistently for every improvement, across every team. That is a data-model problem. A system that records who submitted each improvement, what stage it reached, whether it was implemented, and what impact it produced -- in cost savings, revenue, time, or qualitative categories like safety and quality -- is what makes it possible to ask whether the activity is concentrated in one person or whether the impact came from one lucky idea. The dashboards Roussel showed, tracking activity and impact within a team and visualizing engagement, are the platform's expression of the trust framework.
His three guiding principles -- organized, accurate, disciplined -- are nearly a specification. "Organized" is the case against the Franken-system. "Accurate" depends on data entered consistently rather than re-keyed across mismatched spreadsheets every cycle. And "disciplined" -- the cadence of daily, weekly, monthly, quarterly review, anchored by project impact boards in scheduled meetings -- is what daily-management and dashboard infrastructure is built to support. A dashboard that is always current, visible, and the same across locations is what lets discipline become routine rather than heroic.
And Roussel's standardization advice -- provide people the metrics to track, limit the mind share required, give every location the same dashboard so everyone knows others see it -- is the spread argument. The implementation-rate and impact benchmarks he cited only exist because KaiNexus customers measure the same things in the same way, which is what lets data aggregate into a benchmark at all. A consistent measurement standard, held in shared infrastructure, is what turns one team's reporting into an organization's reporting.
None of this changes Roussel's core message. Start with why. Pick a few metrics. Build the discipline. Listen to your people. Those are choices and habits, not software features. What infrastructure does is remove the friction that makes the discipline hard -- so that getting organized, staying accurate, and reporting on a regular cadence stops being a year-end scramble and becomes simply how the work is tracked.
Why does year-end continuous improvement reporting become a "curse"? Because the data was never organized in one place to begin with. When daily improvements live on a physical Kaizen board, suggestions in a box, A3s and value stream maps handled separately, and everything emailed around as spreadsheets and photos -- replicated across every department -- year-end becomes a frantic exercise in stitching it all together. Jeff Roussel's most striking example: a customer with over 100 locations had each location email a monthly spreadsheet that a central group then had to cobble together just to count ideas. The reporting scramble is a symptom of disconnected systems, not a December problem.
What is the "Franken-system"? Roussel's term (borrowed from KaiNexus partner HubSpot) for what happens when an organization tries to cobble together disconnected tools -- physical boards, suggestion boxes, emails, photos of sticky notes, separate spreadsheets for A3s and value stream maps -- to solve what should be one cohesive problem. The Franken-system is why pulling continuous improvement insights from across an organization either becomes impossible or takes a Herculean effort.
Why does Roussel say to "start with why" before choosing metrics? He framed the session around Simon Sinek's Golden Circle from "Start With Why" -- the principle that you have to establish why something matters before explaining what you will do about it. You cannot answer "how" until you have answered "why," and you cannot answer "what" until you have answered "how." Before selecting reporting metrics, an organization should be clear on why it is doing continuous improvement and why measuring it matters for that organization specifically.
Why does continuous improvement reporting matter? Roussel gave six reasons: we tend to measure what is important to us, so measuring CI signals it matters; you need to know where you are to know where you are going; goals require measurement (the "M" in SMART); the "C" in PDCA means stopping to check, a discipline often applied to work processes but not the improvement process itself; reporting tied to business results keeps executives committed and resources flowing; and reporting helps in both good times (talking about successes) and "bear markets" like leadership changes or waning commitment, where justification keeps the work alive.
What are the three core metrics for continuous improvement reporting? Activity, engagement, and impact. Activity is how much improvement work is happening -- it shows how hard the organization is trying and who cares. Engagement is who is participating and how broadly -- it shows whether improvements come from one person or are spread across the team. Impact is the measurable business value created -- cost savings, revenue, time, and qualitative measures like safety, quality, and satisfaction. Roussel stressed tracking all three: focusing on only one creates blind spots.
Why track all three metrics instead of just one? Because they function as checks and balances. Engagement tells you whether all your activity is actually coming from one place. Activity tells you whether your impact was lucky -- one big improvement -- or the sustained result of a real improvement culture. Tracking all three together is what produces results an organization can genuinely trust, which is what makes better decisions possible. Any single metric in isolation can mislead.
What is a good "implementation rate," and what other benchmarks did Roussel share? The implementation rate -- the percentage of all improvements where the organization actually makes a change -- had held steady between roughly 75 and 85% across KaiNexus customers for three years, sitting at 77% at the time of the webinar. Other benchmarks: about 2.5% of ideas had a financial impact above $10,000, and about 1.4% above $100,000. The implication is that you cannot simply ask for big ideas -- you need volume to produce them. Roussel stressed treating outside benchmarks with a grain of salt, since organizations are at different stages and track different things.
How should an organization benchmark responsibly? Roussel and Graban agreed that comparing only against outside entities can be dangerous and limiting -- the data does not come in a vacuum. Measuring against your own past performance is the more effective long-run path. Graban added a point about who you benchmark: organizations often look for a near-twin (his joke -- a small Vermont dairy producer searching for another with a left-handed CEO). Benchmarking against organizations unlike you -- a health system measuring ideas-per-person against Toyota rather than other hospitals -- compares you against an ideal and can reveal a far higher bar. Roussel also noted that, in KaiNexus data, industry does not predict performance.
How do you get started with CI reporting without being overwhelmed? Roussel's advice was KISS -- keep it simple. People stall by trying to measure 15 metrics for every team and department at once, then losing track of what drives success. Start small: the number of improvements, their financial impact, and the percentage that go to a change is a solid starting trio. Then apply PDCA to the reporting system itself -- create a simple plan, test it for a couple of months, check how both the improvement work and the reporting mechanism are doing, adjust, and repeat.
How do you scale reporting across many departments and locations? Think in terms of standards. A consistent approach to measurement is far easier for people to follow, and since everyone has a day job, you may need to provide the specific metrics they should track (activity, engagement, impact) rather than asking each team to invent its own. Roussel also recommended a visual, consistent dashboard system -- when every location sees a similar dashboard and knows others see it too, rolling reporting out across a larger group becomes significantly easier.
What are Roussel's three guiding principles for CI reporting? Get organized -- stop scattering reporting across mismatched spreadsheets in every department. Be accurate -- without timely, trustworthy data, good decisions are impossible, and inconsistent inputs mean starting from square one every reporting cycle. Develop discipline -- the hardest and probably most important, since many people know how to report and simply do not, because they have not built the regular cadence (daily, weekly, monthly, quarterly) and accountability to do it.
Can activity, engagement, and impact apply outside operations? Yes. Roussel applies the same framework to his own sales organization: activity metrics (calls, emails, meetings, demos) as leading indicators; engagement handled qualitatively by "going to the gemba" and sitting in on sales calls; and impact measured through revenue. The KaiNexus sales team also meets daily to discuss improvement opportunities and tracks them in the platform. As Graban put it, lean is not just for operations -- lean thinking can be applied to everything.
.svg)
Copyright © 2026
Privacy Policy