Impact Analytics and the expensive art of being almost right

Imagine a shop that has plenty of clothes and nothing in your size. The inventory exists. The customer exists. The sale does not. Somewhere between buying the range and filling the rail, a reasonable decision produced an unreasonable result. Retail has an expensive habit of being almost right.

Impact Analytics works in that gap. Its software connects the questions retailers usually answer in separate systems: what will sell, which stores should carry it, how much to send, and when to change the price. The attraction is less time reconciling yesterday’s numbers and more time deciding what to do tomorrow.

A forecast with nowhere to go

A useful forecast needs somewhere to go. DemandSmart builds forecasts down to the product, store and week, incorporating signals such as weather, promotions and price. Different functions can adjust their working versions before an approval process establishes the committed plan. That matters because a marketing assumption and a supply-chain commitment are different things, even when they occupy identical spreadsheet cells.

InventorySmart handles the next decisions: allocation, replenishment and transfers. AssortSmart helps determine which product ranges belong in which store groups. PlanSmart connects the merchandise plan to financial targets and open-to-buy budgets. Together, these tools address a familiar organizational problem: each department can defend its own numbers while the business accumulates the wrong stock.

The spreadsheet has a very good lawyer

Spreadsheets persist because they are flexible, familiar and owned by the people doing the work. Replacing them means replacing a small kingdom of judgments, shortcuts and private knowledge. A technically elegant system can therefore arrive looking, to its intended user, like an elaborate new obligation.

Impact Analytics’ business combines enterprise SaaS with consulting. Its services include data engineering, retail analytics, pricing support and sizing work. The data engineers clean, connect and modernize information systems. Those are commercially significant chores: a recommendation based on inconsistent product hierarchies can be precise about something the merchant never meant.

MondaySmart tackles another everyday ritual, the business review. It tracks retail performance, identifies anomalies and explains possible drivers. Its product description promises visibility into data sources, SQL logic and confidence scores. The interesting ambition is to make the explanation inspectable, rather than asking the buyer to admire an answer delivered from behind a curtain.

A shop that sails away

One company-published case study concerns an unnamed luxury cruise retailer operating across more than 15 cruise lines and 90 ships. A ship is a particularly unforgiving shop. The passenger mix changes; the assortment belongs to a particular voyage; replenishment depends on a port window. Sending stock to the wrong place is harder to repair when the place has departed.

The case describes planning previously dependent on spreadsheets, individual judgment and inconsistent category frameworks. The replacement codified fair-share allocation and built voyage timing into a common workflow. This is a revealing example of the company’s expertise: the relevant unit of planning was not simply a store. It was a moving store with a deadline.

The humans are the implementation

In a public conversation hosted by Impact Analytics, C&A transformation lead Sunny Mall describes an earlier trap in his career: strong models placed over existing processes. Teams kept inspecting every output and making changes. His account concerns previous transformation experience, rather than a disclosed failure of this vendor.

“you have a great model ... but no one’s going to use it”Sunny Mall, C&A - public interview excerpt

What changed his approach was attention to the users and the work itself. He advocates co-creation with planners, business change champions, repeated demonstrations and incremental delivery. At C&A, he describes MondaySmart as an opportunity to deliver value sooner while larger changes continue. The lesson is portable: identify the decision, involve its owner, and simplify the workflow that surrounds it.

Jewelry, jackets and a crowded market

AllSaints announced its partnership in April 2026, aiming to move buying and merchandising away from manual workflows and fragmented spreadsheets. Michael Hill followed in July, selecting forecasting, replenishment, store clustering and assortment tools across Australia, New Zealand and Canada. Its announcement emphasized forecasting performance in high-value, low-volume jewelry categories.

Those are selections and intentions, rather than proof that every promised benefit has arrived. They do show where Impact Analytics fits: enterprise decisions with enough assortment and location complexity to make manual reconciliation costly. Its customers buy software that understands merchandising, alongside help making the operational change.

The market already contains connected planning platforms, including RELEX, which also combines demand, inventory, merchandising and pricing. An integrated suite alone does not settle the comparison. Impact Analytics’ pitch rests on its retail expertise, connected products and delivery support. Buyers must establish that fit against their own data, categories and working habits.

The bill is bigger than the subscription

In January 2024, Impact Analytics announced $40 million in growth financing led by Sageview Capital, with Vistara Growth participating. A May 2025 Series D announcement brought Blue Cloud Ventures into the investor group and named global expansion and agentic AI as priorities. These are company financing events, not customer implementation budgets.

The commercial process runs through sales conversations. A useful proposal needs to specify modules, integration work, data preparation, training and ongoing support. The organizational bill includes time: planners must learn which decisions deserve intervention and which routine recommendations can proceed.

The current agentic platform coordinates specialized agents through Iris, with rules and approvals governing execution. Its practical appeal depends on reliable inputs and an organization willing to change. If every recommendation still requires the old round of manual reconciliation, automation adds another queue. The question worth borrowing from this company is wonderfully ordinary: what work disappears when the new system arrives?