A decision hiding in plain sight

Picture two retail chains. Akkar & Co Group is the giant — hundreds of stores, the biggest revenue in the sector, the name every shopper recognizes. Tripoli Boutique is the smaller player — a regional chain with a loyal but modest footprint, competing against Akkar & Co’s sheer size. Both start the same way: manual inventory counts, paper-based reordering, spreadsheets stitched together by exhausted store managers — no real digital backbone anywhere.

Then both decide it’s time to grow up, picking the same cloud ERP platform and hiring the same systems integrator (SI) to implement it. Same software, same partner, same rough timeline — on paper, Akkar & Co and Tripoli Boutique are starting the race from the exact same line.

But one choice splits their paths. Akkar & Co asks for the system to be heavily customized around its own way of doing things — its own merchandising logic, its own supplier terms, its own store-level workflows. It has a simple conviction behind the request: our processes are the reason we’re winning, and no off-the-shelf system should be allowed to flatten them. Tripoli Boutique, without that same conviction — or the budget to test it — accepts the standard, out-of-the-box configuration the SI recommends.

Three years later, where do these two retailers actually stand — and what does that tell us about the risk of believing your own processes are worth defending?

The fork in the road

Because the software and the implementer are identical, we can rule those out as reasons for whatever happens next. The one real variable is a decision: bend the software to fit the business, or bend the business to fit the software.

Akkar & Co takes the first path, with confidence rather than stubbornness. Leadership had spent years refining exactly how the company merchandises, prices, and replenishes stores — habits and judgment calls that made it the biggest name in the category. “The ERP should adapt to us, not the other way round,” the CIO tells the board, “Because what we do isn’t generic, and we’re not going to let a piece of software pretend otherwise.” So Akkar & Co asks for custom pricing engines, custom store-replenishment logic, custom loyalty integrations — everything tailored to encode what actually makes it good at retail.

Tripoli Boutique takes the second path, and it’s worth being honest about why. It isn’t just budget. Tripoli Boutique has never been entirely sure its own processes are anything special — it has grown steadily, but never by doing anything a competitor couldn’t copy. Standard configuration isn’t a disciplined strategic choice so much as the safer, cheaper default for a company that has never had the confidence, or the resources, to bet that its own way of working is worth preserving in code.

That difference — conviction versus caution — is the whole story.

Two go-lives, one clear fit

Akkar & Co’s go-live is hard. Customization means a longer project, a bigger bill, and more testing when real store data — SKUs, promotions, seasonal pricing — hits a system built to order. Change management is heavy too, because store staff are being asked to fully commit to a tool built around exactly how they already work. But once it stabilizes, the system fits Akkar & Co like a tailored suit, because it was cut from Akkar & Co’s own pattern. Store managers don’t just tolerate it — they recognize their own judgment reflected back at them in how the system behaves.

This isn’t a small footnote. A 2025 report on ERP customization costs found most ERP implementations cost three to four times the original budget, and that modifications made “to improve usability” — the polite name for customization — drive the overspend in roughly two out of three projects. Akkar & Co walks into that statistic with open eyes. For a company confident in what it’s protecting, the extra spend isn’t a risk. It’s the price of keeping an edge intact.

Tripoli Boutique’s go-live, by contrast, is smooth and cheap — and unremarkable. Standard configuration means fewer surprises and a shorter timeline. Staff adjust to the system’s way of doing things because there was never really a “Tripoli Boutique way” strong enough to insist on keeping. It works fine. It just doesn’t do anything for the company that a hundred other retailers on the same platform aren’t already doing.

The AI divide

About a year or two after go-live, the ERP vendor rolls out its next big wave of features — built-in AI: demand forecasting, anomaly detection, automated markdown suggestions, a copilot for store performance questions. This is not a separate project either retailer asked for. It’s simply the next version of the software both of them already own. But what each company gets out of it depends entirely on what it built underneath.

For Tripoli Boutique, the AI is a wash. Because its configuration is standard, the AI slots in cleanly and works well — genuinely useful for demand forecasting and reordering. But it’s the same AI every other retailer running the same standard setup gets. It doesn’t help Tripoli Boutique stand out. It helps Tripoli Boutique keep up, and nothing more.

For Akkar & Co, the AI becomes something else entirely. Because the vendor’s AI learns from the data and workflows underneath it, and Akkar & Co’s data and workflows encode three years of genuinely differentiated retail judgment — pricing logic built around real customer behavior, replenishment rules shaped by real store-level patterns — Akkar & Co’s AI comes out sharper. It doesn’t just forecast demand; it starts predicting the shape of a customer’s experience: what a shopper is likely to want next, when they’re likely to want it, what would make them spend more or come back sooner. The same AI feature, running on top of a genuinely superior foundation, produces a genuinely superior result. This is the payoff the customization was always betting on.

A 2026 industry briefing on ERP and generative AI makes a related point: the retailers getting the most out of AI aren’t necessarily the ones with the newest technology, but the ones whose underlying data and processes were distinctive enough to give the AI something worth learning from in the first place. Generic input, generic output. Distinctive input, distinctive output. Akkar & Co bet on distinctive input three years earlier, without knowing exactly what it would pay off as.

3 years later: the scoreboard

So, three years on — who’s ahead?

Akkar & Co is not just still leading; the lead has grown. That lines up with the broader market too: the biggest ERP vendors and their biggest customers tend to stay on top for a long stretch, and current global ERP market data shows the same concentration pattern, with the largest players controlling a disproportionate share of a market still growing at double digits every year. Size, once earned, tends to keep earning — but this time, faster.

A few things are true at once:

• Momentum was never really at risk. Market leadership built over years — brand trust, customer relationships, scale, supplier power — was always going to be hard for Tripoli Boutique to close in three years. But the ERP decision didn’t just fail to shrink that gap. It widened it.
• The customization was the whole point, not a side bet. Akkar & Co’s custom workflows encoded genuine competitive advantage — pricing logic, replenishment rules, merchandising judgment that actually differentiate the business. When AI arrived, that advantage didn’t just survive the upgrade. It compounded through it.
• Tripoli Boutique’s caution had a ceiling. Lower cost and a smoother go-live were real, honest advantages in year one. But without a genuine edge underneath its systems, Tripoli Boutique had nothing for the AI wave to amplify. Standard configuration kept it safe. It never once made it distinctive.
• The gap that might have narrowed didn’t. In a story where Tripoli Boutique’s agility slowly closes the distance, this isn’t that story. Akkar & Co’s AI-driven differentiation opened a second front Tripoli Boutique had no way to compete on, on top of the first one it was already losing.

Akkar & Co is ahead because of the customization, not in spite of the cost and friction it created going in. The bet on “our processes are superior” turned out to be correct, and correct bets compound.

Image is everything

Once Akkar & Co’s AI started genuinely shaping the customer experience — not just the back office — the company didn’t stay quiet about it. It built a one-year, aggressive marketing campaign around it. Customer-facing stories about AI that seemed to know what a shopper wanted before they did. Public demonstrations of personalized recommendations, faster and more relevant than anything a shopper had seen from a competitor. Messaging that made a simple, confident claim: this is what happens when a retailer’s technology is built around genuinely better judgment.

This wasn’t a defensive move. Akkar & Co wasn’t reacting to a closing gap — there wasn’t one to close. It was amplifying a real advantage it already had. And it worked, for a very ordinary reason: shoppers don’t audit a retailer’s back-end systems, but they do notice when a store seems to understand them better than the last one did, and they talk about it. A 2026 review of branding statistics found that the large majority of marketing executives consider branding essential to business growth, and that companies with consistent, visible branding see measurably higher revenue growth than those without it — branding ROI shows up directly in customer lifetime value and market share, not just recognition. None of this proves marketing alone drove Akkar & Co’s growth — but paired with a genuine capability underneath it, the campaign didn’t need to invent a story. It only had to tell a true one, loudly.

Technology gave Akkar & Co something real to say. Marketing made sure everyone heard it.

The long game

Here’s the part nobody outside Akkar & Co’s boardroom saw coming. Buried in Akkar & Co’s five-year strategic plan, quietly sitting behind the public narrative of “AI leadership,” was a single line item: acquire Tripoli Boutique.

It wasn’t about market share — Tripoli Boutique’s share was modest enough that absorbing it barely moved Akkar & Co’s numbers. What Akkar & Co wanted was something its own mass-market base could never give it: Tripoli Boutique’s customers. Despite its unremarkable technology, Tripoli Boutique had spent 20 years quietly building a small but genuinely valuable clientele — customers with serious money, who bought high-end jewelry and premium cosmetics, and who trusted Tripoli Boutique’s staff enough to keep coming back for exactly that kind of purchase. It was a relationship built the old way, person to person, long before either company touched an ERP system.

Akkar & Co’s leadership recognized that value the moment its own AI proved what it could do with distinctive data. Feed 20 years of high-value jewelry and cosmetics purchase history into an AI engine already sharp enough to predict ordinary shoppers’ behavior, and something bigger becomes possible: a premium tier Akkar & Co could never have built organically, and a much clearer picture of exactly which VIP customers were worth a personal call, a private preview, or a nudge toward the next expensive purchase.

Tripoli Boutique’s board wasn’t blind to the timing. They’d turned down Akkar & Co’s interest twice before, confident their customer relationships were worth more standing alone. This time, watching a competitor’s AI-driven personalization pull further ahead every quarter with no real answer of their own, the board concluded that their most valuable asset — the trust of their VIP customers — would do more good inside a company that could finally act on it at scale than it would sitting inside a smaller one that couldn’t. It wasn’t a surrender. It was a recognition that the asset and the capability to use it well finally belonged in the same building.

The deal wouldn’t be frictionless in real life — a regulator would want a look at combined market share in overlapping regions, and any acquisition of this size rarely closes without scrutiny — but in this telling, the story is about strategy, not scrutiny.

Akkar & Co didn’t just out-execute Tripoli Boutique. It built a genuine edge, told the truth about it loudly, and then bought the one thing its own strength couldn’t manufacture: a customer base built entirely on old-fashioned, human trust.

Closing thought

Akkar & Co’s bet was the right one. Superior processes, encoded deliberately rather than smoothed away by a generic system, turned out to be worth defending — and when AI arrived, that decision didn’t just hold up, it paid out. Tripoli Boutique’s caution wasn’t foolish, exactly. It was simply insufficient: a safe choice that kept the lights on but never gave the company anything to amplify when the moment came.

And here’s the detail that makes the whole plan feel like design, not luck: both companies had been running the same ERP platform all along, just configured differently, so the integration itself was almost frictionless by construction. After the deal closed, Akkar & Co brought the vendor back to the table, not to ask whether the two systems could talk to each other, but to decide how deliberately Tripoli Boutique’s data should be folded in. They chose to merge it fully into Akkar & Co’s customized core, rather than bolt it on as a separate system, precisely so that Tripoli Boutique’s 20 years of purchase history could run directly through Akkar & Co’s optimized processes and AI models — not sit beside them as an afterthought but become part of the same intelligence from day one.

But sit with the ending for a second longer. What Akkar & Co actually acquired wasn’t a smaller retailer’s revenue — it was 20 years of trust that wealthy customers placed in people who knew them by name, and it’s about to run that trust through a machine built to predict what those same customers will want to buy next. The strategy worked exactly as designed. That’s not the uncomfortable part. The uncomfortable part is this: once a relationship built on being known becomes a dataset built on being predicted, is it still the same relationship — and does it matter, to the business or to the customer, if nobody involved can quite say when the answer changed?