When does a spreadsheet stop being enough?
Spreadsheets are the right tool for a business until they are not, and the change is gradual. The signs that you have crossed the line:
- Somebody spends more than 3 hours a week copying numbers from the accounting system, the till or the CRM into a workbook. At 2 people that is 300 hours a year — roughly 8 working weeks.
- The weekly numbers arrive on Wednesday because the workbook takes two days to update.
- There are 4 versions of "sales_report_FINAL" and nobody is sure which one the owner looked at.
- A question like "which customers bought less this quarter than last?" takes a day to answer, so it does not get asked.
- The workbook has become slow, breaks when someone renames a column, and only one person understands it.
- People who should not see margins or salaries can see them, because a spreadsheet cannot hide a column from one reader and show it to another.
Any two of these justify looking at a BI tool. Three or more mean you are already paying for one in wasted hours.
What should a small business connect first?
Connect the system that holds money, then the system that holds customers, then the spreadsheet that holds targets. In practice:
- Accounting — Xero, QuickBooks or whatever you invoice from. This gives you revenue, costs, receivables and cash, which is most of what an owner asks about.
- Sales — the till or online shop (Square, Shopify), or the CRM (HubSpot, Salesforce) for a business that sells through a pipeline. This gives you what is being sold, to whom, by whom.
- Targets and budgets — almost always a Google Sheet or Excel file, and that is fine. Connect it as it is so every dashboard can show actual next to plan.
Stop there for the first month. Support tickets, marketing spend, stock and payroll can follow once the first three dashboards are being read. Klayara connects to 50+ data sources, live to databases or on a schedule from apps and files.
The three dashboards every small business needs
Most small businesses need three pages and nothing more for the first six months.
- Daily trading. Yesterday’s sales against the same day last week and last year, month to date against plan, and a breakdown by product, site or channel. Read every morning in under a minute. Start from the sales overview template.
- Cash and margin. Cash in bank, what is owed to you and by whom (sorted by oldest), what you owe, gross margin this month, and budget vs actual by cost line. Read weekly by the owner and whoever runs finance. Start from the finance snapshot template.
- Customers. New, repeat and lost customers by month, top 20 by revenue, and the ones whose spend has dropped. Read monthly, and before any conversation about growth.
Each page should fit on one screen, show when it was last updated, and have a comparison next to every number. If a fourth page is requested in month one, ask which of the three it replaces.
Should we build it ourselves or have it built?
Drag-and-drop tools mean a capable manager can build the three pages above in a few afternoons. But the person asked to sort out the reporting usually has a full-time job already, and the first dashboard is where most of the decisions about metric definitions get made.
A done-for-you service — where the vendor’s own BI developers connect the sources, define the metrics and build the first pages — is worth considering when:
- nobody in the business has built a dashboard before;
- the data lives in more than two systems that need relating (orders in one, costs in another);
- you want the first pages live within two weeks rather than two months.
To make it work, prepare three things before the first call: a list of the 10 questions the owner asks most often, a note of which systems hold the answers, and a named person who can check the numbers against what they already know. Klayara’s BI developers build the first dashboards as part of getting started; after that, your team edits and extends them.
How do you keep the cost predictable?
Small businesses get caught by BI pricing in three ways: a low entry price that requires a "capacity" or "engine" upgrade once data grows, separate tiers for people who build and people who only look, and charges that scale with how many times a report is viewed.
The simplest model to budget for is one price for every user on a plan, whether they build dashboards or only open the Monday report, with no separate charge for data volume or speed. Then the sum is the number of people who need access multiplied by one figure, and it does not change when you add a second shop or a bigger table.
Two questions to ask any vendor: "what happens to the price when our data doubles?" and "does a viewer cost the same as a builder?" Klayara charges one price per user on each plan and includes the performance for large tables in every plan; the detail is on the pricing page.
Self-hosting for owners who want control
Some owners — often in professional services, healthcare or anyone handling client data — want the reporting platform on their own servers or in their own cloud account, so nothing about their business lives on a vendor’s infrastructure. For a small business this is a legitimate choice, with an honest trade-off.
- What you gain: the data, the dashboards and the AI answers stay inside your boundary; access can be limited to your office network; you decide when upgrades happen.
- What you take on: a server or cloud account to pay for, and someone to apply updates and keep backups — or a vendor who does that with you.
If you have an IT provider already looking after your servers, self-hosting adds little to their workload. If you do not, the vendor cloud is the faster route and you can move later. Klayara runs as the same product either way, in the cloud region you choose or on your own servers, with help for installation and upgrades. See self-hosting and the guide to deciding between the two.
First fortnight, in short: connect accounting and sales, build daily trading and cash, schedule both to the owner’s inbox, then add customers. Everything else waits until those are read every week.