{"id":297,"date":"2026-10-22T18:13:39","date_gmt":"2026-10-22T18:13:39","guid":{"rendered":"https:\/\/wallion.com\/?p=297"},"modified":"2026-10-29T23:41:48","modified_gmt":"2026-10-29T23:41:48","slug":"dejar-la-hoja-de-calculo","status":"publish","type":"post","link":"https:\/\/wallion.com\/en\/blog\/dejar-la-hoja-de-calculo\/","title":{"rendered":"Replacing the spreadsheet"},"content":{"rendered":"
\n

2026-08-31 \u00b7 Wallion<\/p>\n\n\n

\"The<\/figure>\n\n\n

The spreadsheet starts off well. One process, one person, one month ending. The problem isn't Excel. The problem is when the month no longer fits in a single cell.<\/p>\n\n\n

This page isn't the offer. It's the same question as before: is it still an archive or is it the month now? The offer of internal tools<\/a> lives under custom software<\/a>. If the lock is the number, the public table is in How much does custom software cost in 2026?<\/a>.<\/p>\n\n\n

The short answer<\/h2>\n\n\n

A flow.<\/strong> The one that stops work each month when it fails. Budgets, planning, monitoring, or reports. One. Not "digitizing the company.".<\/p>\n\n\n

An owner with a name.<\/strong> Who enters the data? Who sees it? Who responds on the 28th if it doesn't add up? If that's all in "everyone," you don't have a process: you have a chat room with cells.<\/p>\n\n\n

One official number only.<\/strong> Access per person. No two versions of Friday. No reconstructing the data in a separate document to show it.<\/p>\n\n\n

Written scope prior to budgeting.<\/strong> Who, what, what can't fail. Without that page, any tool is just another sheet of paper with better typography.<\/p>\n\n\n

If you recognize two of the signs below, you don't need an essay. You need to decide whether the process remains a file or becomes a tool.<\/p>\n\n\n

Four signs (with the month above)<\/h2>\n\n\n

1. Only one person understands it.<\/strong>
If that person is on sick leave, the month stops. It's not a file. It's a job without a contract. The "I have it in my head" approach doesn't escalate: it's a risk with a structured format.<\/p>\n\n\n

2. There are two versions and nobody knows which one is correct.<\/strong>
The desktop version, the email version, the Friday version. The problem isn't with the formula. It's about who has the latest version. When the official number is in three different places, you don't have a system: you have a weekly fight.<\/p>\n\n\n

3. Every closure requires a fix.<\/strong>
A copy, a paste, a "I'll do this by hand." The makeshift solution becomes part of the ritual. Then no one remembers why. The ritual is the debt: each month more fragile, each month more costly in hours.<\/p>\n\n\n

4. The number you are showing is not the one on the sheet.<\/strong>
You reconstruct it in a separate document because the original sheet can't be shown. If the official data is stored elsewhere, the sheet is no longer the system. It's a draft that someone has doctored.<\/p>\n\n\n

Two signs are enough. Four mean the month now depends on human glue \u2014 and glue takes vacations.<\/p>\n\n\n

What not to do (before ordering)<\/h2>\n\n\n

Don't ask for a system that does everything. "Digitizing" by copying the sheet into an app with the same blanks doesn't fix the month: it disguises it.<\/p>\n\n\n

Don't open a budget without a clear scope. A PDF with a range and no "this is included\/this is excluded" isn't a fixed price. It's a hope with a number.<\/p>\n\n\n

Do not mix internal tools and customer products in the same order. The user of the sheet is usually the team. If the user is an outsider, it's a different table. MVP or internal tool<\/a>.<\/p>\n\n\n

Do not commission "AI on top of Excel" to avoid writing the workflow. Guide: AI-powered automation for SMEs<\/a>.<\/p>\n\n\n

Start with a workflow. Write it down on a page: who enters the data, who sees it, what can't go wrong on the 28th. That's the scope. Without that, there's no tool: there's just another file.<\/p>\n\n\n

How it looks at a small business table (Barcelona \/ Spain)<\/h2>\n\n\n

There's no need for a case study with a logo. The pattern is the table: a founder still signing off on budgets, someone from administration with the spreadsheet that's "almost working," and a partner (consultant, ERP, the A3-type stack next door) quoting the next module when Excel stops working.<\/p>\n\n\n

In Barcelona, the metropolitan area, and the rest of Spain, that group of clients often asks for the same thing: \u201cto stop depending on Maria.\u201d Sometimes the pain is real\u2014a closing that only one person can finalize, a budget spread across three folders, a report rebuilt every Monday. Sometimes it\u2019s about \u201cdigitizing\u201d without a single line of code. Wallion speaks with this type of client often: a concrete workflow; a fixed price; a diagnosis before construction. No fabricated clients. The standard is the client, not a case study.<\/p>\n\n\n

If the sheet is the bridge between two systems that already do the work, it's often best to integrate before building. If the gap is the work\u2014there's no system, only cells and ritual\u2014then you build the flow. Guide: integrate or build<\/a>.<\/p>\n\n\n

Failure modes (those you have already seen, or will see)<\/h2>\n\n\n

Copy the sheet into an app.<\/strong>
Same gaps, better typography. The month still depends on the same workaround. Digitizing isn't migrating cells: it's defining a workflow with an owner and a visible flaw.<\/p>\n\n\n

Budget without scope.<\/strong>
\u201cA tool for managing sales\u201d isn't a goal. It's a wish. Whoever signed off on the issue will either lose it or reopen it with endless change requests.<\/p>\n\n\n

Three flows in a single fixed location.<\/strong>
Budgets + planning + reports + "while we're at it, the CRM." By month two, nobody knows what "done" meant. One workflow. Then the next.<\/p>\n\n\n

Eternal pilot without an owner.<\/strong>
There's an Airtable, a Notion, a script. Nobody's looking at the queue. The problem appears at the end of the month. Visibility first. Then the tool.<\/p>\n\n\n

Compare the flow band with an ERP, an MVP, or third-party brands.<\/strong>
A partner module, a minimum order quantity, and an internal tool are not the same product. Mobility, Glowapp, and Nummo are not this service. Pricing criteria: How much does custom software cost in 2026?<\/a>. Model: fixed price or per hour<\/a>.<\/p>\n\n\n

These errors separate a month that is ending from an archive that is accumulating.<\/p>\n\n\n

What Wallion does (and when not to buy it)<\/h2>\n\n\n

Wallion builds internal tools<\/a> within the same service of custom software<\/a>, to fixed price<\/strong>. It starts with a Diagnostic. A tool\/flow created after the Diagnostic enters the published build pipeline. The license, if we maintain what's been built, is monthly.<\/p>\n\n\n

Model, in a nutshell: a workflow the team actually uses; access per person; a single official number; small changes without reopening the project; someone who answers when things don't add up. It fits when you can write the workflow on one page and you want Diagnosis \u2192 fixed \u2192 license, not a counter "while we digitize.".<\/p>\n\n\n

Don't buy this model<\/strong> When the process is not written down and you want "the tool to discover it"; when the task is to replace the entire ERP in one leap; when the user is a stranger and the signal is from the market (that's MVP, another table); or when you already send a live backlog and need hands \u2014 that's reinforcement \/ staffing: say so.<\/p>\n\n\n

If you say during the call, "this is one flow, not the whole house," it should be mentioned in the first conversation. Geo: \/software\/barcelona<\/a>.<\/p>\n\n\n

Public prices (Option D \u2014 without inventing other figures)<\/h2>\n\n\n

Only the figures that live on the site:<\/p>\n\n\n