A Charlotte florist sold beautifully through Instagram and text messages. The wrong fix was software that would replace all of it. Here's what we built instead.
Most advice for a business running at capacity sounds the same. Get organized. Put it in a system. Automate it. Buy the software, move everything into it, and your growing pains go away.
It is good advice, right up until it quietly breaks the thing that made your business work in the first place.
I think about this every time someone tells a small business that the fix for being busy is more software. Sometimes it is. But a lot of the time the process that looks messy from the outside is not broken at all. It is just informal. And informal is not the same as broken.
City Stems is the clearest example I have of the difference.
A business that was already working
City Stems is a design-forward florist in Charlotte, run by Laura Hughes. Bespoke arrangements and events, the kind of moody editorial work where the flowers genuinely are the design. Her customers find her on Instagram, where she has built an audience of around eleven thousand followers, and they order the way you would expect from a business built on that kind of relationship: they text her. Payment happens over Venmo.
If you have never run a business like this, it is easy to look at "customers text the owner to order" and see a problem to solve. Where is the cart? Where is the order management? How does that scale?
But look closer and the workflow is doing exactly what it should. Instagram is where people fall for the work. Texting is personal, immediate, and completely frictionless: no account to create, no checkout to abandon, just a conversation with the person who is going to make the thing. Laura stays in control of every order and every relationship. For a bespoke florist, that conversation is not overhead. It is the product.
And the channel is not a side hobby. Around 81 percent of Instagram users say they use the platform to discover new products and brands, and 90 percent of accounts follow at least one business (Sprout Social). For a visual, relationship-driven business, Instagram is not a marketing channel that points at the storefront. It is the storefront.
So the real question was never "how do we replace this." It was "what is this workflow leaking, and how do we fix that without touching the parts that work?"
The wrong fix (and why it was tempting)
The obvious move was to give City Stems a "real" website: a full e-commerce storefront with a product catalog, a shopping cart, accounts, checkout, and an order dashboard. Maybe a CRM bolted on to "manage" all those customers.
On paper it sounds like an upgrade. In practice, for this business, it would have been a downgrade wearing a nicer outfit. Here is what it would actually have done:
- Killed the relationship. It would swap a thirty-second text exchange for a checkout funnel. The thing customers loved would become the thing the software removed.
- Forced work that does not exist. A cart needs inventory and SKUs. A bespoke florist does not stock products; every arrangement is custom. You cannot put "whatever is beautiful and in season this week" behind an Add to Cart button.
- Added admin to an owner who is already full. Laura is at capacity. A dashboard to babysit is the last thing a busy owner needs.
- Looked generic. A templated store is off-brand for a florist whose entire pitch is aesthetic. When the way it looks is the sale, "looks like every other Squarespace store" is a real cost.
There is a deeper reason this kind of "upgrade" fails, and it has nothing to do with flowers. Software does not fail because it is badly built. It fails because nobody adopts it. Industry research on business software consistently finds that a large share of rollouts never deliver their intended value, and poor user adoption is the reason more often than any technical fault (Apty). When the tool fights the workflow, people route around it. For a solo owner, "routing around it" means going right back to texting and letting the expensive new system rot.
We have written before about the opposite failure too, the businesses that genuinely have outgrown their patchwork of no-code tools and need something built for real. Both failures come from the same mistake: reaching for a tool before diagnosing the actual problem.
What we built instead
We start every project with a question that is less exciting than "what should we build" but far more useful: what already works here, and how do we get out of its way?
For City Stems, that meant building a storefront that acts as a front door to the workflow she already runs, not a replacement for it:
- It looks like her. The site carries the same moody, editorial aesthetic as her feed. It feels like an extension of the Instagram people already followed her for, not a jarring detour into generic-store-land.
- It keeps the portfolio live. Her Instagram feed is embedded directly, so the work stays current without her double-posting anywhere.
- It hands the visitor to the conversation. Text-to-order calls to action run throughout, and on a phone they open a pre-addressed text message with a tap. The site's whole job is to route a stranger into the same conversation Laura already has with her regulars.
- It keeps payment where it was. Venmo, with a QR code on desktop and a direct link on mobile. No checkout to rebuild, no new fees, no change for the customer.
- It catches the form-fillers too. For visitors who would rather type than text, a contact form routes straight to her phone.
Here is what actually changed. Before, every Instagram visitor who did not already have Laura's number hit a dead end. The demand was there; the path to act on it was not. Now the site catches those people and funnels them into the text flow that already worked. She did not change how she runs her business. She stopped leaking the demand her Instagram was already generating.
Replace the workflow vs. complement it
| Replace the workflow | Complement the workflow | |
|---|---|---|
| Starting question | What system should they be on? | What already works, and what is it leaking? |
| Customer experience | New accounts, carts, checkout | The same easy text they already send |
| Owner's new workload | A dashboard to manage | Nothing new to learn |
| Brand fit | Whatever the template looks like | Built to match the business |
| Risk | Low adoption, quiet abandonment | The workflow keeps running, now with less leakage |
Neither column is always right. The point is that the column you belong in is a diagnosis, not a default.
What Laura said
The proof that this was the right call is not a metric. It is that the owner barely had to lift a finger, and the result looked like her.
"Hurrah Web Solutions is phenomenal! Mike made building my floral design website process remarkably easy from start to finish. As someone who isn't technical, I loved being able to trust him completely and stay mostly hands-off while he handled every detail. He created a gorgeous website that truly represents my brand and included all of the features I wanted without making anything feel complicated. Mike's communication was fantastic, via text or email, he is incredibly professional, responsive, and made the whole experience stress-free. I'm so grateful for Hurrah's talent and expertise!"
Laura Hughes, owner, City Stems
"A gorgeous website that truly represents my brand." "Trust him completely and stay mostly hands-off." "Made the whole experience stress-free." For an owner at capacity, that last one is the whole game. The project added zero to her plate and gave her back the visitors she had been losing.
The principle worth stealing
You do not need to run a flower shop to use this. Before you buy or build anything, run two checks:
- Is the process actually broken, or just informal? Messy-looking is not the same as broken. If the informal version is fast, keeps your customers close, and you can run it in your sleep, be very careful about "systematizing" it into something slower.
- After launch, will you do less, or will you have a new system to feed? Good software removes friction from what already works. If the tool leaves you maintaining it instead of the other way around, it is working against you.
This is the entire reason we build custom software instead of dropping clients into a template. A template makes you adopt its workflow. A custom build adopts yours. If you want the longer version of how we think about that trade-off, we laid it out on our comparison page.
And to be clear, "complement" is not code for "never replace." Some processes are genuinely broken. When EZ Cell was running a growing wholesale operation out of spreadsheets and DMs, that was a real bottleneck, and we replaced it with a custom app that gave them 15+ hours a week back. The difference was the diagnosis. EZ Cell had a broken process. City Stems had a working one that was leaking at the edges. Different problems, different builds.
The mistake is skipping the diagnosis and reaching for the same "get it into a system" answer for both.
If this sounds like your business
If you are at capacity and every "just get organized" pitch feels like it would break how you actually run, that is worth a conversation before you sign up for anything. That is exactly the kind of thing we untangle in a free thirty-minute audit: what is working, what is leaking, and whether the honest answer is "build something," "complement what you have," or "you are fine, do not let anyone sell you a dashboard."
Not ready to book? Tell us the one task that eats your day and we'll send back a free sketch of what we would automate first (and, just as often, what we would leave alone). Or see how we compare to agencies, no-code, and freelancers.
