Someone went looking for a trash can. At McGuckin Hardware, the Boulder retailer, the website returned nothing. This was a vocabulary problem with commercial consequences: the words a customer typed did not connect with the words in the catalog. In Unilog’s telling, its software suggested alternatives. A human could review them. A dead end became a piece of maintenance someone could actually finish.
- Unilog sells commerce software and the product content that makes it useful.
- Its customers include distributors, manufacturers, and specialty retailers with complicated catalogs.
- The practical lesson: examine what buyers cannot find before buying another feature.
It is a modest opening for a technology story. No new shopping ritual. No reinvention of retail. Just the difficulty of making a computer recognize something a store employee would understand immediately. But that difficulty is a useful way into Unilog, a business whose history begins well before the current enthusiasm for artificial intelligence.
The catalog comes first
Unilog began in 1998 with cataloging services. Its founder, Achutha Bachalli, came from industrial management. A 2014 YourStory account describes a business hit hard by the dot-com downturn: a workforce of roughly 100 shrank to 15. Survival preceded the software platform. So did the patient work of organizing other companies’ information.
By 2012, the argument for changing the business was becoming explicit. Selling labor cheaply offered little protection from the next cheaper supplier. Suchit Bachalli, now Unilog’s CEO, put the problem bluntly when discussing hourly rates with The Economic Times.
“Even if you reduce it from $20 to $8 an hour, there is no entry barrier.”Suchit Bachalli · The Economic Times · 2012
The newspaper reported that Achutha had started with ₹200,000 of personal savings. Unilog dates its flagship B2B commerce platform launch to 2014. The sequence matters: the company learned the catalog problem before selling the storefront. Product data became part of its proposition, alongside software that could turn that information into a transaction.
A storefront with a memory
Consider what a distributor asks its website to remember. A customer may have a negotiated price. Stock may sit at several branches. A purchase may require approval. The person ordering may need a quote before anyone can buy. Each detail is ordinary inside a distribution business, and potentially awkward inside a shopping cart built around a simple retail transaction.
Unilog’s configurable CX1 CIMM2 platform addresses those workflows with account tools, requests for quotation, approval rules, and connections to enterprise resource planning systems. CX1 Connect supplies the integration layer. It links the digital storefront with the systems holding operational information, including pricing, inventory, and orders. That connection is central to whether the site can answer a buyer’s question.
Then comes the merchandise itself. CX1 PIM organizes product information and publishing. Content subscriptions supply enriched descriptions, specifications, and imagery. Showroom and SPEX Builder products support discovery, presentations, and specification documents. Essentials targets a simpler starting point. These products fit different selling situations, from an industrial reorder to a customer comparing bathroom finishes.
Product content
PIM
Commerce + ERP
Useful information travels all the way to the buying decision.
The maintenance bill
Unilog occupies a specialized corner of enterprise commerce. Buyers can also consider platforms such as Optimizely Configured Commerce, BigCommerce, or Adobe Commerce. Unilog’s distinctive proposition is the combination of commerce, managed product content, and distribution expertise. The company currently advertises more than 18 million actively managed product SKUs and over 400 customers across North America.
Its business model combines software subscriptions, content subscriptions, and professional services. For a buyer, the sensible cost comparison includes implementation, catalog preparation, integrations, and continuing upkeep. A platform that saves time at checkout can still demand substantial work behind the scenes. The purchase decision should count that work. A useful buying exercise is to follow one real order from the product record to the invoice. Who supplies the specifications? Which system knows the available stock? Who fixes the mismatch when the two disagree? Those questions make a software proposal much easier to judge.
Capital helped Unilog extend the proposition. Kalaari Capital invested in 2015. Investcorp announced a majority acquisition in January 2021, with terms undisclosed. Bachalli later explained that the attraction included Investcorp’s ability to handle a company operating across India and the United States. Product development, support, and adjacent acquisitions were his stated priorities. Bravo Business Media joined Unilog that year, expanding its showroom capabilities.
From a backlog to a workflow
The newer AI products address another cost: the work that accumulates after launch. McGuckin’s blog production stalled after it lost the employee doing the writing. Search maintenance competed for scarce staff time. Unilog’s case study describes embedded Blog and Synonym agents helping restart those routines, with staff reviewing suggested search terms inside the existing commerce workflow.
Customer-reported example published by Unilog. A task comparison, rather than a promised return.
In August 2026, Unilog launched CX1 Sigma, bringing commerce, content, and its HyperScale Growth Agent Suite into a new platform. Its announced Site Studio creates pages from plain-language instructions or a reference URL. Unilog says agent actions are previewed and recorded before publication. The direction is clear: give a lean commerce team more execution capacity, while retaining a review step.
Start with the failed search
There is a useful practice here that does not require buying Unilog: read the searches that return nothing. Compare customer language with catalog language. Check whether a product record contains enough information to make a decision. Assign someone to fix the recurring gaps. Those are concrete jobs, with visible before-and-after states.
The limits follow from the same logic. An integration still depends on trustworthy operational records. AI suggestions still need someone who understands the merchandise. If no team owns product accuracy, automating publication can spread the confusion. Unilog’s appeal is strongest where a business has a complex assortment, established buying rules, and recurring information work. The sale starts when a buyer can recognize the right thing.