The six-step process you described sounds similar to a standard data governance plan that many large corporations already have. Is there a meaningful difference, or are you essentially advocating for data governance with a dashboard?

Yes, that is essentially what I am advocating for. In smaller companies, data governance and a formal plan often do not exist at all, so the six-step method gives those teams a practical starting point. In larger corporations, the governance plan may exist but the research results often do not get connected to it, so the findings end up not being implemented or measured. The gap I see most often in larger organizations is not that they lack research, it is that they lack the infrastructure to carry the research through to action and measurement.

How do you get leadership buy-in when you want to build a dashboard or data governance process and it requires significant resources?

Start by pulling together whatever data you already have access to and building a case with it. Show leadership the specific business problem, quantify it where you can, and then explain what you need to go further. If you do not yet have permission to access all the relevant data, have the conversation first and ask for access to run the analysis. Frame it as a problem-solving pitch rather than a resource request. In my experience, leaders who understand the problem are generally willing to open up the systems needed to investigate it.

What do you do when an individual contributor does not have access to all the data sources they need?

Start with the data from your own department and build out the plan as far as you can. Then present that partial view to leadership alongside a clear articulation of what additional access or resources you need to complete it. Whether the company brings in an external consultant or promotes someone internally to own the project, leadership will ultimately need to grant access. The goal is to make the problem concrete enough that they understand why access matters.

Internal Operations: How startups learn from their own data & when to call in research

Julia Schwan, founder of Opero, draws on a decade of finance and operations experience across startups and large corporations to present a six-step method for aggregating internal data before deciding whether to commission external research. Using a SaaS membership renewal problem as a running case study, she shows how disconnected tools and inconsistent data identifiers obscure business problems, and how centralizing that data produces both cleaner decisions and stronger briefs for research partners. The talk is equally relevant to lean startups with limited budgets and to larger organizations that receive research results but lack the data infrastructure to act on them.

Key Takeaways

Disconnected tools and inconsistent naming conventions are the root cause of most internal data problems. Standardizing unique identifiers across systems is a necessary first step before any analysis is meaningful.
A six-step method, from naming the problem through assigning owners, scoping, building, cleaning, and acting, applies to both lean startups pulling data together for the first time and large corporations trying to measure the impact of research they have already received.
Clean data is the byproduct of a clean process. Aggregating data will surface broken workflows, missing contacts, and duplicate records that must be fixed before conclusions can be trusted.
The decision to bring in external research should be proportionate to company size and revenue. A company doing $2 million a year may be better served by calling clients directly, while one doing $80 million has more reason to engage a research firm.
Having centralized, clean data before engaging a research partner produces a more precise brief, faster leadership buy-in, and a clear baseline metric, making the research investment more likely to generate a measurable return.
Large organizations often have research but lack the data governance infrastructure to implement findings or measure their impact. The same six-step process can be run in reverse to deploy research results effectively.

Questions & Answers

The six-step process you described sounds similar to a standard data governance plan that many large corporations already have. Is there a meaningful difference, or are you essentially advocating for data governance with a dashboard?: Yes, that is essentially what I am advocating for. In smaller companies, data governance and a formal plan often do not exist at all, so the six-step method gives those teams a practical starting point. In larger corporations, the governance plan may exist but the research results often do not get connected to it, so the findings end up not being implemented or measured. The gap I see most often in larger organizations is not that they lack research, it is that they lack the infrastructure to carry the research through to action and measurement.
How do you get leadership buy-in when you want to build a dashboard or data governance process and it requires significant resources?: Start by pulling together whatever data you already have access to and building a case with it. Show leadership the specific business problem, quantify it where you can, and then explain what you need to go further. If you do not yet have permission to access all the relevant data, have the conversation first and ask for access to run the analysis. Frame it as a problem-solving pitch rather than a resource request. In my experience, leaders who understand the problem are generally willing to open up the systems needed to investigate it.
What do you do when an individual contributor does not have access to all the data sources they need?: Start with the data from your own department and build out the plan as far as you can. Then present that partial view to leadership alongside a clear articulation of what additional access or resources you need to complete it. Whether the company brings in an external consultant or promotes someone internally to own the project, leadership will ultimately need to grant access. The goal is to make the problem concrete enough that they understand why access matters.

Session Notes

Context and Speaker Background

Julia Schwan is the founder of Opero, a firm that connects disparate software tools so companies can see their data in one place. Her background spans a decade in finance and operations, including a role as controller for the office of the CEO at SAP (then roughly 100,000 employees), a finance leadership role at Jump for 50 Media (a 25-person marketing agency acquired by Omnicom Media Group), a co-founded SaaS startup called Scrubs that helped international doctors obtain U.S. medical licensure, and a COO/CFO role at Training and eTracking Solutions, a five-person online training company. This range, from Fortune 500 to sub-ten-person startup, shapes the dual perspective she brings to the talk.

The Core Problem: Data Lives in Silos

Most companies, regardless of size, operate with data spread across tools that do not communicate with each other. Common examples include:

Finance: QuickBooks, Xero, SAP
Marketing and CRM: Mailchimp, HubSpot, Zendesk
Project management: Jira, Asana, Monday.com
HR and payroll: Gusto, ADP, Justworks, Rippling
Collaboration: Slack, Teams, Gmail, Outlook
AI tools: ChatGPT, Claude, Gemini, Copilot

A compounding issue is the unique identifier problem. Different people in different departments set up the same client under slightly different names. One person enters "Acme Corp," another enters "Acme Inc.," a third uses "Acme old client." When someone tries to pull a unified view of that client's data, the records do not match. This problem grows proportionally with team size.

The data stays disconnected for three main reasons:

No one has it as a full-time responsibility to coordinate across tools.
Teams within their own function often do not know something is broken elsewhere.
There is no clear owner assigned to the problem.

A Concrete Example: SaaS Renewal Failures at Scrubs

At Scrubs, the company was selling recurring memberships to international medical students. After a strong first year of acquisition, renewal rates started dropping. The data needed to understand why already existed inside the company, but it was scattered across five separate places:

Product usage data (platform logs)
Sales notes (kept separately from the CRM)
CRM contact records (some with outdated decision-maker emails)
Renewal dates and contract values (QuickBooks)
Support tickets (handled only by the support team, not shared)

Once aggregated, the combined view surfaced a clear pattern. A client renewing in 28 days had used 15 of 100 purchased seats, had not been contacted in 74 days, had a new VP whose email was unknown, and had filed nine support tickets in the past 30 days. No single team had seen all of that at once. By the time it was visible, there was little time to save the relationship.

At the time that I'm looking at this, there's very little chance that I'm going to salvage this client in 28 days.

The Six-Step Method

This method applies in two directions: bottom-up for startups identifying a problem with their own data, and top-down for larger organizations trying to implement and measure the results of external research.

Name the problem. Be specific. "We are losing renewals" is more useful than a vague sense that something is wrong. Ask why repeatedly to get past surface symptoms.
Assign the people. Identify a project owner who understands the business problem and a builder who can connect the systems. These are often two different people because the skill sets rarely overlap. In a five-person company they may be the same person.
Scope the project and define the data fields. Determine exactly which data points could answer the question. For the renewal problem: client name, revenue, margin, renewal date, last sales touch, last support contact, decision-maker contact, and product utilization. Agree on definitions before building anything.
Build and pull the data into one place. Smaller companies (under roughly $1 million revenue) often do this in spreadsheets. Larger ones use BI tools such as Tableau. Today, internal teams can also build lightweight dashboards using AI tools (Claude, ChatGPT with artifacts, Cursor) connected via APIs and deployed on platforms such as AWS or Vercel.
Fix the problems and processes you find. Aggregating data will expose gaps: empty fields, duplicate client names, multiple revenue streams counted differently across systems, processes that exist only in someone's head. Resolving these is not optional. Bad input produces bad output.
Make a decision and act. With clean, connected data, prioritize the highest-risk segments first, for example, high-revenue clients with low product utilization. Then decide whether your internal data and team are sufficient to address the root cause, or whether external research or consulting support is warranted.

When to Call in External Research

The six-step process does not replace research. It creates the conditions under which research becomes more useful and more efficient. Indicators that it is time to bring in external help:

Your internal data shows what is happening but cannot explain why.
You want to test a new concept and need structured methodology and objectivity.
Your team lacks the time or expertise to conduct rigorous inquiry at scale.
Company revenue justifies the investment. At roughly $2 million, direct client outreach may be sufficient. At $80 million, a formal research engagement is more appropriate.

The value of completing the six steps first is that you arrive at a research partner with a precise question, a clean customer list, agreed definitions, a baseline metric, and a quantified sense of the problem's scope. That produces a better brief, faster internal buy-in, and a clearer way to measure the research's return.

The Large-Company Angle: Deploying Research Results

In larger organizations, research typically arrives top-down: leadership commissions a study, receives findings, and needs to cascade an implementation plan to teams. The same six-step method, run in the opposite direction, provides the infrastructure to do that. Without a connected data environment and clear ownership, research findings tend not to be implemented effectively, and there is no way to measure whether the business actually changed as a result.

You have the results, but then how do you measure the ROI of that research? How do you measure if your team is actually implementing what the research is recommending and if your business is actually turning around?

Getting Leadership Buy-In

For practitioners who need to advocate internally for a data governance or dashboard initiative:

Build the case with the data you already have access to, then present it as evidence of the problem.
Frame the ask around business impact, not just ROI. Relevant angles include client satisfaction, team capacity constraints that are blocking new revenue, and retention risk.
If you lack access to all data sources, start with your own department's data and present that partial view alongside a clear statement of what additional access you need.
Reframe the request as a sales pitch: this is the problem, this is why it matters, this is how I plan to solve it, and this is what I need from you to move forward.

Transcript

Read the full transcript

Thank you so much everyone for uh for continuing to hang in there. My name is Luke Williams. I'm part of the accelerant research team and uh I'll be helping kind of facilitate our next presentation which will beginning be beginning shortly. Uh just wanted to uh to cover a few ground rules before we get started. We are at the mercy of technology. Uh, so if someone has an issue, uh, an audio issue, uh, we can't see the Q&A, something like that happens, you know, please, uh, please hang in there with us and we'll get things sorted. I think we've all been in that position before. Uh, please be respectful. Uh, we do would like to foster an active conversation, but um, you know, just make sure that, uh, you are engaging with an information need in in a very positive way. Sometimes when we're not face to face, it's easy to be unintentionally brusk. Uh so we'd like to try to avoid that here. And um really feel free. This is an opportunity to network, engage, and interact. The digital channel is a little different from doing that in person, but uh hey, we are bravely marching into a new world. Uh so if you have any questions or would like to have a discussion feel free feel free to use that Q&A channel and we will uh we'll we'll try to uh get that to the presenter and get a response. All right, our next presenter is going to be Julia Schwan. She is the founder of Opero and her presentation will title is internal operations how startups learn from their own data and when to call in research. Julia, let me know when you're ready to go and uh you can take over the presentation screen. >> I'm ready. Thanks, Luke. Let me go ahead and share my screen. Hey, everyone. Good afternoon. And then Luke, would you mind letting me know if there's any comments in the chat at any point? >> Absolutely. Uh do you want me to uh bring them up as they occur or save them for the end? Um, let's do as they occur. Let's do interactive. >> Okay, we'll do uh not only am I monitoring the current Q&A, but we are live streaming via YouTube. I'll shout out any questions that come across there as well. Thank you. >> Okay, that's perfect. Um, all right. Nice meeting you everyone. My name is Julia Schwan. I am the founder of Opero. And as Brett and Luke mentioned, I'm actually not a traditional researcher. I don't conduct research myself but I am on the other side of companies that received research and I have been working mostly on startups and when Brett and Accelerin reached out to me uh the idea was to bring the perspective of how do internal operations work in smaller companies lean companies where we might not always have the budget and the resources to immediately allocate into research and what do you do from there like how do we aggregate data? How do we learn from our own data and if you are not from a startup and you you work in a much larger corporation there is a a piece here that I do think it's also important because we are going going to be talking really about um how do we aggregate data in a day-to-day basis in a corporation so we can also plug in that research. So if you receive the results of research that you're conducting or you hired an external researcher to perform where do you go from there right like where do you plug the results to cascade that down to your team to see the results to see um the ROI of that decision. So um before we start I am going to be introducing a little bit more about myself and why I am speaking about this topic but also feel free to place a chat uh question. I can see the chat here as well but and Luke's going to be um flagging any questions. I would love for this to be a conversation not only me speaking for about an hour um and yeah we can go from there. So to start just a little bit about myself so you know where I'm coming from. I've been working in finance and operations for about 10 years. I started my career at SAP. SAP is the one and only in the screen that is a big Fortune 500 company at a time that had 100,000 people working. And that was my really my experience of how big organizations run. how are decisions being made and how results of research um cascade down to the teams. At the time I was the controller for the office of the CEO and we used to get a lot of data from external parties and consulting agencies on how we would cascade that down to our team and that was the angle that I observed um from a large corporation. I'm very very curious to see if that's still how it works when we get to that part from you all talking on the chat. But from there I had my first experience in a smaller company. I joined Jump for 50 Media. They are a marketing agency. So primarily media buying in New York City. And at that time I was employee number 25. So it was my first time seeing how a small corporation runs and what do you do when resources are a little tight and where budgets are not super large. Um I was brought in to really build the finance department at Jump. So I built the department. I was not only the person doing the budget and the forecasting but I was the payroll manager, the HR manager. Um wearing a lot of hats and that is something that really excited me. I led Jump's acquisition to Omnicom Media Group. Um, so I stayed there for a while and after we went through the M&A and Jump was acquired, I actually took a first step at starting my own company and I started Scrubs community. Scrubs was a company where we helped international doctors to become doctors in the USA. So I am originally from Brazil. That's the accent. And my partner is a doctor in Boston. She has gone through the process of becoming a doctor in the US. And if there's any nurses, I think Brad told me that there's hospitals and people in the medical field here. You're familiar with the USMLE process. We were teaching international doctors how to become doctors in the US. So she, my partner, was teaching the medical part. I was running the company in terms of everything else, sales people, hiring, firing, really trying to figure it out how we would monetize. Um, and that was the first time I was really trying to pull data together internally in my own company to make better decisions like how can I bring all the data that I have, how can I build processes in this company so I can make a better decision as the owner of the company. From there I became the COO/ CFO of training and etracking solutions. That company um is a really cool project. They specialize in online training for people with mental disabilities and any other types of disabilities. Um uh that was a five people team. So very similar to scrubs. I was really focusing on pulling data from all the sources that we had in a company centralizing it so I could make my job faster and more effective and I could have a clear picture of what was happening in terms of when we came to a problem. what is even my baseline to try to solve this problem. Where's the data? How can I see numbers? Um, and that really is what led me to found Opero. Opero is exactly that. I make softer talk for companies. I pull data together uh from different tools that a company might already use and they tell me, Julia, I am not interested in purchasing any new software. I just need the current ones that I already used and work for my company to talk to each other so I can see my data clearly and I can make a better decision. So that's what I do now and this is where I come in to talk to you guys from this perspective of how can we create processes and pull data to make internal better decisions from the startup level when you don't have enough resources and from a big company level when you do have the re results of the research where do you plug that how do you actually cascade that down to your team so that's what we're going to be talking about Today um in terms of an agenda I will be starting with the two points. So what is the startup point and what is the large Fortune 500 company? Where do they start and what do I see? I would love your feedback there. If that's also where you see things starting what is the problem which is usually data connection. We have all this data internally in companies but really how do you connect that? The bulk of our hour is going to be in what is the six-step method that I like to use to organize this flow and to bring that clear vision to bring your data together so you can then make a decision when to call research from the experience working in those companies when um I would call research when I wouldn't call research and then of course a Q&A. So my goal is if it goes well in this hour, you would leave this um webinar presentation understanding how you can connect your internal data to learn from it. That's the main goal and it would help smaller companies startup style because I'll be sharing what is the process that I developed and in all those companies that I worked for um how do I personally gather data so I can make informed decisions and al again from a large company perspective where do you plug the results? So starting with the two points large and small companies on the large company side what I tend to see is research comes top down. So there is a problem that is identified. There is internal research happening in the company either a internal department that performs research or there's a consultant firm that is hired and brought in to understand a problem. The company leadership receives the results of this research and then they need to cascade down an implementation plan. and when they need to cascade down the implementation plan, teams need a connected view of that information to then be able to plug in that research. On the other side, in companies that are doing between 1 and 20 million, that's usually the the the frame that I see. Um, sometimes in that level of revenue bringing in research, it's not an option. So how do those companies which is my field of expertise how do we internal internally pull data together to be able to see what where what what is the problem where is coming from and then make a decision on what to do with it. So what is the connection between the two of them is the one connected view is what's missing. So then you get to disconnect the data and why things are missing. And personally, has anyone in the the chat here, let me try to open the chat. Are you familiar with any of those tools? And if yes, do you mind putting in the chat which ones have you used before? Um, let's see if this chat is working. No. Does anyone mind saying hi? Just so I know that the chat's working. Okay, perfect. Thanks, Chris. Um, all right. So, if you haven't, if you're not familiar with any of those tools, um, there are many departments that use separate tools in a company, right? So, in the finance perspective, again, coming back from like a small perspective of company, QuickBooks is usually the accounting system that it's used. Zero is also very popular. SAP has a accounting system as well. All of those have all all those angles and um boxes have many other options. Marketing, Mailchimp and HubSpot are really uh popular for sales and CRM. I would say that HubSpot is very familiar. Zenesk as well. And then you have to deliver um in a project management tool. So, Jura, it's a tool that is widely utilized. Azana Monday.com you come to your people's perspective and payroll you have gusto you have ADP you have just works you have your PO and then everyone collaborates in either teams or slack and email Gmail outlook and then we now have AI as a a component as well that your company might be operating on chat GPT claw gemini co-pilot right so this is one of the reasons for this problem is that data lives in all those different tools that normally are not really talking to each other. And a separate layer of that problem is that whoever is setting up the unique identifier and the unique identifier in this case could be the client name, right? So the Acme client, the person setting up Acme in all those different software and tools might utilize a different unique identifier. And I know that because I've done it myself as being a single person creating the unique identifier in multiple tools. I've created different names for the same client. Imagine when you have five different people for each department creating the unique identifier. So you end up with things like Acme Corp, Acme.com, Acme Inc., Acme old client, Acme new client and that doesn't really help your data person when they try to pull the data together or your operations person to then understand a compilation of that view. So then why does this stay disconnected? Right? So like all the data is living in all those different silos and tools. um usually is because nobody really has time to to coordinate all this. If it's no one's full-time role to perform that, uh it's really difficult to find someone who's just going to take time out of their day to connect all this. Sometimes people don't have the knowledge that something is broken to begin with. So, the person in accounting only takes care of the accounting and the finances. They don't really know that there's something in monday.com or any of the project management tools that is creating a mismatch. And the real gap that I see it's ownership. So is identifying the problem and then creating a role a person that would be the owner of that project the owner of the the tool coordination to then solve a problem and bring your data together again with the idea that there's a problem that you're trying to solve and you either have to bring research or try to figure it out internally um with less resources. So then this is a an example. I'm a very very visual person. So an example to um showcase what I mean by that. And this is an example of my own company. So scrubs. So I don't target any of the other companies. Um at scrubs we were pretty much selling SAS software as a service. We were selling a reoccurring membership into a a community into an education portal that we had. And the first year went great. We we're we're doing really well. I was really happy with the operations. And then I started noticing that we weren't renewing a whole lot of clients. And the purpose of SAS and the purpose of memberships is that you will renew clients at some point, right? You spend so much time getting the leads, putting your sales team to work, convincing someone to join your cause and join your uh product and service. And the renewing is supposed to be the easy part. But I realized that a lot of our students weren't renewing with us when normally they would. And then I realized that all the data that we needed was already existent inside of the company. it was just not aggregated. It wasn't together in a way that I could look at it and flag that something was going on and I had to make a decision or I had to act on it. So an example is that the usage of that client was in one platform which was the platform that we would serve them. We had notes from the last sales touch that had happened that was in a notes um platform that the sales team had their own notes about things that wasn't combined with the CRM for some reason. We had the CR CRM with the email of the decision maker who by the way had already left that company that was our client and he wasn't the contact anymore so I couldn't contact anyone. We had invoices with the renewal dates and the amount of that client, how much that was worth for us in QuickBooks in our finance department. And then we had the support chat with a lot of frustration from clients that only the people in support were handling separately from all those other areas. So each team had a view. No one had really put things together side by side until we had the problem of we are losing clients, we need to take a look into it. So once I finally did pull everything together, I got to something that looks like this, right? So, it's a more visual way to get the data that is scattered across all those tools and have a red flag saying this client that is about to renew in 28 days has used 15 out of the 100 seats that they bought. They didn't have a touch with us in 74 days. They have a new VP. We don't have the email from that person. And they had nine separate tickets in the last 30 days. they might be frustrated with you. At the time that I'm looking at this, there's very little chance that I'm going to savage this client in 28 days. And that's what we started seeing. And this is what triggered me to be very passionate about overall data in a company which then you use to create research. So with that problem, this is the method that I started utilizing in every other company that I started working. So the first one, the first step is to really name the problem. And sometimes we think that it's easy to name what's really happening. And I was watching one of the panelists before, I think the 11 a.m., and she was um talking about asking the why and the why and the why and the why. um you really need to do some brainstorming, understand what is really the problem that we're trying to to identify and do research on. Right? So then you assign the people responsible. Easier when you have a company with five people and you are all the people together. You are the entire team and you can put everything together. A little harder if you have a 100 people and you have 200 or maybe thousands of people and heads of each of those departments. You really need to identify who are the responsible people for each of those areas. The third step is the actual starting of the build of what you're going to pull together. It's the scope and the definition of what's going to be created and built. You're going to pull things in one place, which is step four. You're going to fix the problems and the processes because you'd see that once you start getting data, you will find gaps in your data. you will find the mismatch, the double the names, the the information that wasn't properly allocated. And if you have a bad input, you're going to have a bad output. And then you're going to make a decision and act. And this is the part where for the smaller companies who are doing from the bottom up, you decide if you want and need and can bring external research to solve the problem. and you do the the opposite way if you already have the results of the research and you use the six-step process to then um organize your systems, organize your flows to then be able to measure the results of that research and the impact and the changes that are happening in your company. So let's go one by one. So, in keeping the SAS example that I used before, I'll keep that same example throughout the presentation um because I think that's easier for us to relate back. My problem is I'm losing renewals. Nobody's renewing with me. I worked really hard for a whole year to get clients and I'm losing all of them. That's the problem. Who are the roles and the people that I'm going to sign? What I tend to see is that we need a project owner. We usually it's a project manager. It's someone who's going to take this project from zero to one to one to 10 and is going to continue monitoring. So in the case of the dashboard that I showed before, who's actually going to look at that data? Because once you have a dashboard and nobody logs in and nobody continues putting data into it, it's just a waste of time. Um, so who's going to be that person? And who's going to be the builder? And in again a larger corporation that could be multiple owners and multiple builders. I'm going to treat it as like one and two so it's easier for the the concept of the presentation. But who's really going to connect it all? And usually tends to be two separate people because the knowledge of the building tends not to be the knowledge of understanding what is the problem of we are losing clients. So once you identify the two people or the teams that you have, you can start moving into scoping the problem. The third one is scope the project, define the words, define everything that you need to solve the problem to answer the question that you have of we are losing clients. Why? What are the information? What are the data points inside of our systems that could give us an answer to why we're losing clients? At this point, we can just put a lot of pe a lot of words on a whiteboard and think it's bad customer service, it's bad technology, it's bad um content, the actual product that we're selling is not good enough. There's a million reasons, but we don't really have any data backing that. So in the example of we are losing clients, it's the client name on a list, revenue that a client is bringing and I would say margin is as well because a client could bring you a lot of revenue and there's so much cost involving servicing that client that the margin of the client is actually not grandiose and it might not be a worthwhile client for you. So margin would be also an option. What is the expiration date of this renewal? When was my last touch? When did my customer support team talk to this person? Um, who's the decision maker? In that case, the VP that I was talking to no longer works for the company and I don't have a backup email for that person. That doesn't help me much. And then in this case, what is the utilization of them? How much are they using the product or service that I am selling? So you have the fields, you have the definition, you have the cadence of when you'll bring that data and again this is between the person who understands the problem which tends to be the COO, the project manager, it could be the CEO in the case of the smaller companies that are out there and who is building this and it could be the same person. It could be someone like me that I had five people I had to figure it out and I started building things. Um, normally I see it being two different people. So once you get to that point, it's when you start to build and I was watching as well the niece speak about AI and AI is a big theme in every every conversation that I've been having lately. So, I did bring examples of how to build this with AI because there's no reason for someone internally in your company not to pull that data together if your company allows and it's um um adding AI to their systems. However, there is a same there is a possibility that you do all this with spreadsheets and Excel and AI is not a must to have that step four built. So if companies are in the mark of like less than a million dollars in revenue, I see a lot of spreadsheets happening. If the company is on the larger side, this is where companies used to use tools like Tableau, uh BI tools and dashboards. And nowadays, the very exciting thing is that internal teams can build this with AI and cloud and artifacts. Um, cursor is a builder as well that you can toggle multiple types of AI to then pull all the API keys that you need. You would deploy this in a Amazon AWS or Versel or something similar and you have the weekly cadence. So I don't want to spend the majority of the time talking about AI since it's more about the process and not the build. But I did want to bring that it is a very exciting time that most companies can build this for themselves and this is available to all of us to then be able to digest the data and improve our internal processes even before we bring in research. Once you get to that step and once you start doing that, this is where you'd see the data starts breaking because you'd see empty fields. You'd see that there's three or four definitions for the same customer, which is that example that I gave with Acme. You'd see that there's two different versions of revenue. You have one bucket in your accounting system that is overall revenue, but you have B2C and B2B, and you have renewals, and there's more buckets for that same item. uh you'd see that processes many times just live in someone's head um someone's notepad someone's internal disk in their computer and it's not in the cloud is not being shared and we need to make the discovery and bring the processes that have been working into everyone's mind um and there's accounts that nobody remembers opening that was a case in many a few of the companies that I worked for we were it was a 20 to 30year company. We had 10,000 clients and some of the clients weren't even working with us, but nobody had marked them as terminated or turnurned. So, we had more CL we thought we had more clients than we actually had for that reason because nothing was really integrated. So, clean data is the byproduct of a clean process. Um, I'm going to pause to see if there's anything on the chat. Okay, cool. Uh yes, there are things. Let's see. Hi Julia, I know lots of these tools. Thank you for saying that. I've been using Slack, Jres and I'm just seeing this now. Thank you. Uh I have experience at Husb Monday.com plus everything. So everybody's like understanding what I'm talking about. This is really great. Thank you. Uh losing clients, trouble with renewals, maybe your software sucks. Is crappy software every day. Uh that is a possibility. Yes. Uh, okay. So, no real questions, but I'm glad you know that I'm not speaking a different language here and most people have been using this. Um, three definitions of customers in two versions of revenue. Yes. So Jacqueline what I meant by that is when you start aggregating data an example is you had um I had a slide in the beginning that the unique identifier the name for that client was type different in a few of the systems right so you have to um create a conversion for the unique identifier of that client an example is if you are familiar with QuickBooks and you enter the name of that client in QuickBooks as Opero LLC But then in a different system it's just oper and in a different system it's not it's um it's written all caps. So there's different ways that teams call different customers especially when you have like 20 30 40,000 clients. That's what I meant. Um and then two versions of revenue. You could have a mainstream of B2C clients, a mainstream of B2B clients, a mainstream of just re renewals. And you have revenue coming from different systems. So you can have revenue coming from Stripe, you can have revenue coming from like an invoice that you send out to the client, you can have revenue that is coming from uh Square. So it's just different systems that you have different source of revenues all coming to your books. Um thanks for the questions. I appreciate the the interaction. So then once you get to that step um and things are breaking, this is where you're going going to start fixing problems. You're going to identify issues that you have in the company and you're going to start fixing this problems which is going to be key for then you get to the research and like go to a research with like an actual clean data set or if you already have the research to start deploying that and seeing the results of this with a clean data set. So this is on the middle of that process and then with that you would get to something that just visually is an example but you you will get to this you get to an example where you pull the data from all those sources. Your data is clean. You figure out the definition of all the clients and unique identifiers and all the systems that you need to pull from. So then you have a visualization or you have an internal tool that all the key stakeholders in your company can look and then understand what's happening in this case with the clients like why am I losing clients because nobody's using the platform because we had nine tickets of support they were unhappy with us nobody replied um they have a new point of contact that we don't want didn't identify so this is where you bring the problems into surface is and you can act on it and then as a step six you can actually decide what to do. So in this example I would call or like I would pay attention first to the clients that bring a lot of revenue and have a low utilization on my platform. So clients that are not using what I'm selling my product or service and they have high revenue so they're important to me but probably they're not going to renew because they're not using it. So that's the first group that I'm going to take action and I am going to call I'm going to sit down and call those clients and say hey I understand that you're not using the platform as much. Um why is that? Would you mind explaining a little bit more? And then you're going to start learning about your clients. And this is the point that you might decide to then bring in research because you started with the problem of I am not renewing clients. You went through six steps to identify what's happening. This was just to see your data in a clean way and understand what's happening. And right now you understand that there are clients that pay you a lot of money and they are not using your product. And then you get into why. So you can set up calls with all of them. You can email all of them. You can get in contact with them. You can set up processes to get in touch with all your clients more often that you have been. Then you can decide that your own data stops there and you don't have the team, you don't have the resources, you don't have the experience and you prefer to bring in experts to help. This is where historically I've brought in research and consultants to uh companies that had that power or then I brought in a team and started doing my internal research with the resources that I had at the time. Um because you need to start identifying what why and how why are the clients not renewing in ex in this example creating a turn analysis. Um you do want to try test a new concept. This is the time that you can test a new concept with experts guiding you through the way. And what as you probably know, bringing an expert saves you time that you might not have in a lean operation with five and 10 people um personnel and the experience. However, what it's really going to determine is the size that you're dealing with. So, if you're bringing in 2 million a year, maybe just call the client yourselves. Maybe you don't have the size yet and the resources to bring external experts. If your company is bringing $80 million a year, maybe call in and research, maybe ask for some help to identify the problem. But regardless, once you have your data centralized, you'll be able to bring a precise question, you'll be able to bring clean data and a customer list that works. You would have agree definition because on step two you already spoke with your team and you decided who are the champions for this project, who's the builder and who's the decision owner and who's going to carry the torch of that project. You'd have a baseline metric. You would understand what are the clients that are renewing, what are the clients that are not renewing. And you have a size of that population of I have 20% of my business that is in risk because of this problem. I didn't realize this before. let's now act on it and understand really what's happening. So if you decide to bring or not bring research internally, you would have better briefs a faster buyin. So if you are the one, if you're the operator like I I used to be and I am bringing this to leadership, you would actually bring data and saying I've already pulled the data from all the sources in a company. My job stops here like I cannot go further. Could we go externally and hire help to identify what the problem is and how to course correct it? That's a much faster buy in then we've been losing clients. What what yes >> thank you. Uh there's a comment in chat and I'm going to paraphrase it a bit but the the it's about the predeide to bring in research or not process that you've been walking through. Um it it seems a little familiar to a standard data governance plan that many large corporations have. It are are there nuances to what you're talking about or is that that you can differentiate them or is that really what you're advocating for uh a data governance plan with dashboard and all those tools? >> Yes, it is. It is what I'm advocating for. So uh thanks Christopher. So in in my world in a world of smaller companies data governance and a plan many times is just not existent. So for companies like that using the resources internally and following a plan of data governance is what's going to bring you to the results of yes let's bring experts or no like now let's identify what is the problem and this is what I'm calling internal research in much larger corporations what I've seen and I love your perspective Christopher as well is you do have research but then maybe you don't have that data governance plan in place to carry the research and plug in the research. So you have the results, but then how do you measure the the ROI of that research? How do you measure if your team is actually implementing what the research is recommending and if your business is actually turning around? U I would love to hear from you because my that's my experience is once you get the results, the data is still messy and then the research tends to die. many many times. You're welcome. Yeah. Uh but yeah, if you if you can type, I'd love to hear from you like if that is actually what happens in larger companies or that's just my experience and I'll come back to that. Uh but yeah, I appreciate you you putting things on the chat. That's that's awesome. Um and then all right so the recap and I might have actually sped through this and have a lot of time but the the recap here is what I see it's it's really important for us to connect the data that we have so that data governance and fix the problems that you find for sure once you start connecting your data in terms of mismatches revenue you different sources, processes that are broken. And then once you have that clean data is when you get into doing research or choosing if you'd like to bring an external party of research, choosing if you'd like to hire someone um to conduct that research inside of your company. Or if you already have a research plan, you would move on the opposite way and create that data plan and clean clean up the data to be able to deploy the research. And again, you guys talked about researching different aspects of the business. This example was client acquisition, client retention, more business focused. Um, but I would love to hear your thoughts in terms of if you've experienced this, if this it's something that in different areas of research you also used. Um, and yeah, open to any questions. >> Thank you, Julia. Uh, we have one from YouTube. One second. Golly, it's lot. Uh, so, um, basically, uh, LEV, uh, I like the idea of what you call it, dashboard of governance. uh I'm on board uh advocating it to leadership. That's a really big setup. How do you get buy in if you have a specific decision area in mind? Um it's a lot of resources. So a mini estimate case like ROI brought to leadership uh is a is a clear one. Do you have any other tips? >> Uh let me see if I understand. So she he or she's asking how to get buyin from leadership when you want to conduct research and you want to bring someone externally >> uh when it when you want to put together a dashboard or a governance process like you were walking through with the uh renewal piece. >> Okay. Yeah. Um in in my experience is it's a it's a conversation is I I will I put the case. So I used to lead finance and I would have a problem that maybe my my leadership wouldn't think it was a problem, right? But in my my opinion that was a big problem. I would put a case in a sort of presentation. I'll pull all this data like I would create the dashboards or I'll pull the data that I have available to me to then explain my point of view. And if it's not return on investment, is it client satisfaction? Is it us not being able to like close new clients because our team is too tied up which is something that I used to face as well. The team is too tied up on the current clients. We cannot go get new revenue and all the pressure that I was feeling in terms of being a CFO and head of finances revenue is not growing. But I had this other thing tying us up. So I would pull that together before and then take that to get the buy in. If the question is, do I even have permission to get that data to pull it together to then get the buyin? I would just have a conversation with them first and say, hey, this is what I'm trying to do. Uh, especially in a small company, I'm assuming you're talking about a small company. This is what I'm trying to do. This is how I am planning to pull that data. I need access to X, Y, and Z to then present a plan and then go from there. But Luke, let me know if this answers the question, please, since I don't see YouTube. >> We'll do. Cheers. I see no other open uh questions in the Q&A channel. So, everyone uh feel free to drop those directly into the webinar chat. Thank you. >> And Chris, I see that you commented. Thank you. I find that you're correct with smaller companies. I also find that they're immature when it comes to data management. Exactly. And that's where um a lot of people like me and accelerate research comes in. But I also found that what happened to that business that you founded? Um, my partner became a a top researcher at Harvard and we decided that that would become a fun business for us. So now it's a passion project, but she uh is dedicating her time doing that and we decided to do it together. So this this is now a passion project. We still have students, but it's not a business anymore. Okay. Uh Lev commented in the YouTube channel. Uh question answered. Thank you. Uh advocate to leadership on board. Uh we have uh we have a couple comments in the chat if you want to take a look at them or I'd be happy to read them out if not. >> Okay. No, I'm I'm happy that this this has been useful. Uh thanks for uh Ivonne. I also said thank you Julia. This is help was helpful. I am leading a division of a medium-sized overseas company. I need to pull the data to make my case to build our business. Yeah, especially overseas. So the I I that company that I was running was happen in Brazil as well. And then you start dealing with um data and API keys from other sources and it it can get tricky. But I would say is if your company is AI forward, there are ways to do that nowadays that could could help you bring your case to leadership. What about when an individual doesn't have access to all the data sources which are silent? Yes. Uh that becomes harder and that's where you would build your case and then take to leadership. So um let me know what what is the size of your company more or less. But I'm thinking about a company that has 100 people and you are maybe like at a third degree level from leadership. I would bring that topic up in conversations and explain what you're trying to solve and what are the the keys and resources and people that you need to pull that together. In all those cases, leadership has to buy in on your idea. If not, you never get access to everything that you need to even pull the data. I'll start with the data of your department. I'll start with the plan and saying this is this is how far I can go internally. this is where I would need your help or if I cannot get access to this, you might want to bring an external consultant, external research firm to solve this problem. The difference is that regardless, they would have to open up the systems and the keys and connections if they're bringing someone externally or if they're hiring or promoting someone as the champ of the team to do it. They just need to understand the problem. And this is kind of like a sales pitch, right? like this is a problem. This is why this is a problem. This is how I'm trying to solve. And I find that every time that this happens and people bring those to me, I'm actually very excited as a company uh leader and I I open whatever they need. But I understand that it sometimes it's it's a difficult situation. Um thanks Rod. I appreciate that. Okay, while we're waiting for any closing questions, thank you again, Julia. Any final thoughts for you or summary or questions you were hoping or expected to be asked that you wanted to answer that did not get asked? >> And also, this is my personal email. Uh, sometimes you don't want to ask questions, but if you do have um yeah, just shoot me an email. I'll be happy to to talk to you guys. And thank you uh Brad Bill Accelerant Research for the invite. Um I appreciate the space. >> Okay, thank you everyone. Really appreciate you joining the conference. Uh I believe there's another one scheduled for next month's. So uh please be on the lookout via email or check the conference site and sign up. Uh several of the decks were requested during the presentations. uh we will collect those from the presenters and those will be archived on accelerantressearch.com site. Uh you can either go directly to the subsite link that was in your invitation today or log on to the main site and type conference in the search bar and that'll take you uh right to the archived presentations when they're available. Thank you so much everyone and thank you again Julia. Excellent presentation. >> Thank you. Have a good one. I guess.