Prices_FINAL_v3.xlsx: What Manual Price Collection Really Costs

Day one: your category manager asks a merchandising assistant to check competitor prices on the top 40 SKUs before Thursday's trading meeting. Day three: the assistant has 40 prices, gathered from 15 open browser tabs across four competitor sites, pasted into Prices.xlsx. Day five: a second assistant, unaware the first list exists, starts her own version because the original file "looked out of date." Day eight: someone notices the two files disagree on 11 SKUs and can't work out which is right, because neither file records when a price was checked. Day eleven: Prices_FINAL_v3.xlsx goes into the trading meeting — and the prices in it are already six days old.

Nobody in this story is being lazy. Six people touched that spreadsheet: the category manager, two merchandising assistants, a buyer who added his own tab for "extra context," a manager who reformatted it for the meeting, and an analyst who tried to reconcile the conflicting versions. Four email threads carried updates, corrections, and "which version is current?" questions. Eleven days elapsed between the original ask and a number anyone could act on. This is not a hypothetical — it's the default shape of manual competitor price collection at retailers who haven't automated it, and it's worth being honest about what it actually costs.

The Process Isn't the Problem — It's the System

The instinct is to blame the spreadsheet. It isn't the spreadsheet's fault. Excel is a fine tool for analysis. It's a poor tool for collection, and the two get conflated because collection is where a spreadsheet workflow starts.

A manual price collection process has three structural weaknesses that no amount of discipline fixes:

No single source of truth. The moment a second person opens a copy to "just check one thing," you have two files with the same name and different contents. Version control by filename (_v2, _FINAL, _FINAL_v3) is not version control — it's a record of how many times the truth diverged.

No timestamp discipline. A price without a capture time is a price you can't trust. Was it checked this morning or last Tuesday? Most manual trackers don't record this, because typing a timestamp for every cell is exactly the kind of tedious step that gets skipped under deadline pressure — which is most of the time.

No error visibility. If someone mistypes €89.90 as €809.90, or copies the wrong SKU's row, nothing flags it. The number sits in the spreadsheet looking exactly as authoritative as every correct one, until someone catches it by eye — usually the person who has to explain the anomaly in the meeting.

Where the Eleven Days Actually Go

Walking the timeline above against what each step really involves:

Add it up and the fortnight cost roughly two to three person-days of direct labor — spread thin enough across six people that no single line item on anyone's calendar looks alarming. That's exactly why it survives budget review: the cost is real but diffuse, hidden inside "admin time" rather than showing up as a line item anyone questions.

The Hidden Costs That Don't Show Up in Hours

Time spent is the visible cost. Three more costs matter just as much and rarely get counted:

Decisions made on stale data. By day 11, the competitor may have already moved again. A pricing decision built on six-day-old numbers is a decision built on a market that no longer exists. A 1% improvement in pricing discipline correlates with an 8–11% increase in operating profit according to McKinsey's analysis of S&P 1500 companies — and stale inputs are one of the fastest ways to erode that discipline before a decision is even made.

Coverage that shrinks under pressure. Forty SKUs is already a small sample against a catalog of thousands. Manual processes don't scale down gracefully to "cover everything" — they scale down to "cover what we have time for," which means the long tail of your catalog is priced on guesswork, not data, indefinitely.

Institutional knowledge that walks out the door. When the assistant who knows which competitor site is fussy about being scraped by hand, or which SKU mapping is unreliable, leaves the company, that knowledge leaves with them. Nothing about a manual process is documented well enough to hand off cleanly.

What Automating Collection Actually Changes

The fix isn't "work faster" — a faster human still has all three structural weaknesses above. The fix is separating collection from analysis: let a system capture prices continuously and consistently, and let your category manager spend their time on the decision, not the data-gathering.

An automated feed changes the shape of the same story: prices for the same 40 SKUs — or 4,000 — are captured on a schedule, timestamped automatically, matched to the correct product, and delivered as one dataset with one version. There's no second assistant duplicating work because there's no ambiguity about where the current numbers live. The competitor price tracking layer becomes infrastructure instead of a recurring fire drill.

This doesn't mean every retailer needs an enterprise pricing platform on day one. It means the collection step — the part currently eating six people's time across eleven days — is the part worth automating first, independent of whatever repricing or analysis tooling you build on top of it later. If you're weighing build vs. buy for that step, see our breakdown of in-house vs. managed price monitoring costs.

How ScrapeWise Fits

ScrapeWise replaces the manual collection step with a scheduled data feed: competitor prices captured daily or hourly, matched to your SKUs, timestamped, and delivered as one clean dataset — no tab-hopping, no duplicate files, no version conflicts. Category managers get one number they can trust instead of a spreadsheet they have to audit before they can use it.

Honest limitations: ScrapeWise handles the collection layer — it doesn't replace your repricing logic or merchandising judgment, and matching accuracy still depends on clean product identifiers on your side (EAN/GTIN, brand + model) for the cleanest results. Pricing is based on SKU and competitor volume rather than a flat self-serve rate.

Conclusion

The eleven-day fortnight above isn't a failure of effort — it's the predictable output of a process with no single source of truth, no timestamp discipline, and no error visibility. Fix the collection layer and the rest of the pricing process gets faster by default, because everyone downstream is finally working from the same, current number.

Tired of reconciling _FINAL_v3 files? Book a call →