It's time to demand more out of your software partners and become more self-reliant.
Many of you, I'm sure, remember the old JG Wentworth commercials from about 15 years ago. And if you don't, here's the link.
As a marketer, I loved those commercials. I loved the catchiness and the brand recognition that came from a group of people yelling "it's my money and I need it now."
As a consumer, it drove me nuts. You'd leave the room at halftime of the game you were watching only to hear people start screaming from your living room, which always caused a pause or a "this commercial is so stupid" moment to roll through your head.
Now, I'm not here to talk about financial services or settling debt. I'm here to talk about the parallel between that commercial and where software and data sit today, because it is high time we all start screaming "it's my data and I need it now."
Years and years ago, before AI, the cloud, and the SaaS boom, companies relied on self-built internal systems and onsite servers. That server was a physical machine sitting in a closet or a back room of your own building. You bought it, somebody in house kept it running, and you paid for that capacity whether you used it or not. Files lived on a shared drive on that box, version control was whatever your team agreed to that week, and files traveled around as email attachments. That is exactly how you end up with five copies of the same budget and nobody quite sure which one is right.
I still remember one of my first marketing jobs, where I was responsible for managing the weekly ad for a group of independent grocery stores. I quite literally printed 8-10 copies of that week's creative proof and distributed them every morning to the desks of all the buyers around the office. They'd mark up corrections by hand throughout the day. Then I'd come collect them in the afternoon, scan them in, and type out numerous emails back and forth with the design team to make the changes needed, all while meticulously keeping track of versions.
Take that kind of process and duplicate it across every department. Finance, HR, Operations, and everyone else all had their own systems and their own ways of manually keeping things organized. There was an Excel spreadsheet or a Word document for EVERYTHING.
Then along came the cloud, which really just means the software runs on somebody else's servers instead of the one in your closet. Google and Microsoft turned documents and spreadsheets into something several people could work in at once. Salesforce proved a serious business system could run in a browser, HubSpot and others followed on the marketing and CRM side, Oracle and Intuit moved their products to the same model, and platforms like Monday.com and Asana made real-time collaboration normal.
And for a while, all of it felt like it was starting to work. The silos came down, a mountain of manual busywork went away, and for the first time everybody in the building could look at the same information on the same day. But a tradeoff came with it that most of us didn't notice at the time. Your data now lives inside somebody else's system, on their terms.
Somewhere along the way, most companies adopted more software than they can count. What started as an awesome solution to manual work and communication inefficiency has quickly evolved into a huge headache. Most organizations now face a tangled web of tools that are expensive, probably not being used for their intended purpose, and all trying to become the "all in one" solution for everything.
And now we have AI entering the mix. Every software stack is bolting on chat features and assistants that often make things more confusing than helpful (minus a few that do it well), and on top of that you're hearing all about new companies like Anthropic (Claude) and OpenAI (ChatGPT) and trying to figure out where they fit with everything you already own.
What I keep hearing from clients, partners, and folks in my network is some version of the same thing. I just want something that works, with easy visibility into data that doesn't cost an arm and a leg in software, IT, and development. Something I can actually access and make good business decisions off of.
In short, "it's my data and I need it now."
I've been skeptical of self-built systems for a long time. The premise behind them was always fairly sound, because no matter how good the software companies were, they could never do exactly what a company needed. But building it yourself meant development cost, ongoing upkeep, no real API integration, and clunky reporting. So it always looked like a choice: commit to the software and use it the way the vendor intended, or go build your own thing and take on all the headaches that come with it.
Here is what I had wrong. I don't think that's the choice anymore.
Yes, your company is unique. Everybody's is. But "we're unique" was never really the interesting part, because everybody says it. The more useful truth is what's sitting underneath it. That spreadsheet your ops leader refuses to give up? It usually exists for a reason. There was a question the software couldn't answer, a report it couldn't produce, or a number that needed to be looked at a way the vendor never anticipated.
So let me say this plainly. It's OK to keep it in a spreadsheet. Seriously.
Running outside software alongside your own internal sheets and tools is not a discipline problem, and you shouldn't feel bad about it. Every leadership team I work with has at least one person keeping "my sheet" next to the system of record, and it is usually the most trusted number in the building. The software is good at the one or two things it was built to do. The spreadsheet fills the gap it left. Neither one is the problem. The problem is when the two can't talk to each other.
What you should expect in 2026 is the ability to get your information out of those systems and bring it together in a way that actually reflects how your business runs, quickly and cheaply enough to make a decision on a Tuesday instead of waiting on a month-end report.
My other skepticism was AI itself. I've sat through a lot of demos and heard a lot of promises, and for a long time my read was that AI was going to be one more expensive thing companies buy and try to use that never really solves the problem. It just polishes a turd when the data and processes underneath it are already bad. After the developments of the last 6-12 months and quite a bit of testing, that opinion has softened.
Part of the problem is still how AI gets talked about. The conversation is usually about what it might eventually become, or the same demo everybody has already seen, which is writing a social post or cleaning up an email. Meanwhile the things I'm actually watching change businesses are a lot less glamorous. Reading a messy export. Lining up two data sets that were never built to talk to each other. Doing in a few minutes the spreadsheet gymnastics that used to eat somebody's entire Thursday.
That changes the equation, and it changes it most for small and mid-sized companies. For years, when a business couldn't get the visibility it needed, the answer was to buy another piece of software, upgrade the one you had, pay for an integration, or hire someone to build something. AI hasn't made any of those go away. What it's done is dramatically lower the cost of filling the gaps between the systems you already own.
Which means the first question doesn't have to be "what software do we need?" anymore. The better question is "what decision are we actually trying to make, and where does that data already live today?" Most of the time you don't have a data problem. You have an access problem. The information already exists. It's just sitting in three systems that were never designed to talk to each other.
If you run a business somewhere between $1 and $50 million, you already know the moment I'm talking about. You ask a straightforward question in a Monday meeting and it takes four days and three people to answer it. Your controller has one version of the numbers, your ops lead has another, and marketing is reporting results that don't tie to either. Nobody is wrong and nobody is hiding anything. The data exists. It's just sitting in four systems that were never built to talk to each other, so getting to it in the shape you need costs more time than the decision is worth.
Here's my own example. I have no coding experience. Working with Claude, in a very short amount of time, I built an HTML dashboard that pulls our Databox marketing reports, QuickBooks financial data, and Monday.com operations data into one view. It filters, it's fully editable, and it shows the business the way an owner needs to see it instead of the way three separate vendors decided to show it. No developer, no crazy integrations, no API costs.
Three sources, one view, no developer.
And it doesn't have to be a full dashboard. Between clients, partners, and my own network, the builds that change things the most are usually smaller than people expect.
None of those required throwing out the software already in place or buying another platform. That's really the point. Use the software for what it does well. When there's a gap, you don't automatically have to spend more. In a lot of cases, filling that gap just got a whole lot easier and cheaper.
Now, I'm not the guy who is going to tell you to run your entire company on AI and vibe coding, or to throw out the software you just spent two years implementing. Start with the need. What decision are you trying to make, what number do you need to make it, and where does that number live today? Pick the one question that costs you the most time every month, build the smallest version of the answer, use it for a month, and then add to it if it actually helps.
And none of this works without process and people. A dashboard built on bad inputs is just a faster way to be wrong, and somebody on your team still has to own pulling the data and keeping it accurate. Larger enterprises obviously carry another layer of security, compliance, and IT requirements that make this a different conversation, but the same systems thinking still applies. The technology is changing fast. The need for good judgment isn't.
With AI and the right processes in place, there are fewer and fewer excuses for not having visibility into your own data. The answer isn't that everyone needs to go build their own software, and it isn't adding another platform every time you hit a limitation. For most small and mid-sized businesses it's somewhere in the middle. Use outside software where it makes sense, keep the internal tools and spreadsheets that are actually useful, and expect the pieces to work together.
Think about your last leadership meeting. What simple question could nobody answer without saying "let me get back to you"? Start there. The answer probably already exists somewhere in your business. If it lives in a tool you're paying for, you should be able to easily get to it in 2026. And if you can't, it may be time to start demanding more from that partner, looking at other options, or building the missing piece yourself.
Need help figuring out where to start or what the right mix is? Reach out to us today. Between our team here at GCS and our partners across disciplines, we can help guide you through it.