How Wood Sales is moving 60 years of surplus inventory onto one system

Sixty years of surplus and closeouts

Wood Sales Company has operated in Golden, Mississippi since 1968. Owner Tracy Chambers' parents started it in general merchandise and grew it into importing tools; today, in her words, “surplus and closeouts is the basis of what we do.” The company also runs customer service for First Alert safes, including a replacement-key inventory, and sells through several Shopify stores, eBay, Amazon, TikTok, EDI orders and its own online store, Toolmart Inc (TMI).

For most of its history the business ran on paper. “We're almost 60 years old, and until the last five years we've had no computer system at all,” Chambers said. The software that did arrive came in layers — an Access database for the showroom, QuickBooks Desktop later, some inventory in Shopify — and the company never fully moved onto any of them. Convexity is the first system meant to hold all of it.

Three systems and a paper warehouse

Mike Timms wears many hats at Wood Sales — sales, inventory management, IT, a call center — after close to 30 years in information technology. The operation he inherited runs inventory in three places at once: an Access database that still serves the showroom, QuickBooks Desktop that the company never fully moved onto, and Shopify for part of the catalog. “Not everyone knows that certain things are in certain places.”

Underneath the systems sits paper. Pick, pack and ship is largely done by hand, every outgoing order is recounted by a second person, and QuickBooks pushes the rest of the work into Excel: it cannot hold the fixed retail, wholesale and dealer prices Wood Sales needs, and “you can't get very granular when you're doing reports because QuickBooks has limitations on everything.”

The excess stock and forecasting problem

For a surplus and closeout business, the inventory is the business: it is bought opportunistically, in bulk, and it only earns money if someone remembers to sell it. That is exactly what the current tools fail at. “It will allow us to move excess inventory that we have,” Chambers said of Convexity. “Sometimes you forget about it when it gets stuck in the corner and you don't have good tools to see what you do have in stock.”

The cost compounds across teams. The sales team cannot see what is available to sell. Purchasing cannot see what is stagnating, so it rebuys items similar to bulk stock the company already holds. And nobody can say which warehouse locations are empty when merchandise needs to move.

Forecasting is the other half of her case. With customer buying history in one place, the sales team can work the list instead of their memory: “We can see what our customers have bought in the past and tell them, ‘It's been a while since you bought this,’ or have something similar in the line that we can possibly add to their line as well.”

What one record changes for an operator

Timms' case for accuracy is a story every operator recognizes. “If my inventory system shows I have three of these and a customer wants 100, I say, ‘I have three right now.’ Then I walk out and I see a box with 200 in it. I've lost a customer, more than likely, because he's gone to the next guy who's keeping an accurate inventory.”

The key room shows the other failure mode. Two customer-service staff sold First Alert replacement keys from a Shopify count; when it hit zero but keys were still on the shelf, they kept selling. New keys were ordered but never added to the system, so several keys now show negative seven or negative eight on hand — and the true count “could be off by hundreds of keys either way,” from nothing but human error.

With every record in one system, the count is the same one everywhere it is read. His target is not perfection: “Having a 100% accurate inventory, I know 100%'s not always the case, but within 2% either way … is critical in this business.” Within 2%, an operator can trust the screen instead of walking the floor.

Scanning instead of recounting

Today, every outgoing order is counted twice. “Somebody ordered 100 of these and 50 of these, April calls me, and I have to walk downstairs and recount what she's already counted. And there are still mistakes even with two people checking it.”

Convexity's warehouse app moves location transfers, receiving and pick, pack and ship onto a handheld scanner. Each item is checked off as it goes on the pallet, so the second count — and the walk downstairs — disappears: “If every item is checked off and put on the pallet, there should be little to no error from that point moving forward.”

Change management, in his telling, is a sentence: “There is no old system. Now we're on Convexity.” After decades of parallel systems — some staff have run the same process for 30 years — the point of the rollout is that there is exactly one place to look.

What she wants AI to take on

During Convexity's onsite visit, the team photographed a handwritten piece of paper and had Convexity's AI create a new vendor and a purchase order from it. For a company that still does “lots of things handwritten,” that demonstration landed: “Yesterday we were able, from a piece of paper, to tell it to create a vendor in the computer and to also create a purchase order.”

The first big job she wants AI to take on is the First Alert key room: decades of handwritten key records that staff search by hand while a customer waits on the phone. “Nobody wants to tackle that task. Everybody's been putting that on the back burner 'cause it's a large job.” Digitizing it means faster answers for customers who are already upset when they call.

She is just as interested in AI as an interface. Staff who have never worked in a computer system can chat with it instead of learning screens: they can work in it “without actually having to know the ins and outs.” In a workforce where some employees have run the same paper process for 30 years, that is what makes one system adoptable at all.

Getting the time back

The payoff he cares about is time. “I'm the most excited about alleviating some of this manual work, because it's not hard, but it's time-consuming. Time is your most critical asset. You can make more money. You cannot make more time.”

His estimate: “four to five different people would save around an hour and a half a day. So five and a half hours a day. That adds up quickly.” Asked whether better forecasting would also cut excess inventory and stockouts, he answered, “Oh, absolutely.”

On AI, he is an enthusiast with standards — he uses it daily and has no patience for AI used as a shortcut: “Use AI to improve your life, not to literally make you lazy.”

The payoff: sales, profit and time

Chambers declined to guess at hours saved — “Let's answer that after we get going. I think it's gonna be a lot, though” — because hours are not what she is buying. Her list is commercial: the sales team knows what it has to sell, old inventory gets seen and moved instead of written off, and customer purchase history keeps the right products in front of the right customers.

“I think sales and growing profit are the key things that we will get from Convexity.”