What data capturing should cost — per record, per hour, per project

Most data-capture quotes collapse the moment you dig into them. A provider says "R15 a record" and you think you understand the cost. Then cleanup turns into a separate line item, accuracy review takes weeks, and a "simple" field mapping balloons into custom formatting. The real lesson is that pricing models don't tell you what you're actually paying for—or what work is assumed away.

Why hourly rates hide the real commitment

When a data capturer quotes you an hourly rate, they're typically covering their fingers on the keyboard. What they're not quoting is the mental load. Deciphering handwriting, interpreting abbreviations, deciding what goes where when a form is ambiguous, or flagging records that don't fit the schema—these decisions slow down the keystroke count. If you're paying only for "logged hours," you're inviting a provider to rush through judgment calls or to start billing you extra for every interpretive judgment. The hourly model also obscures whether they're working from clean source documents or wrestling with PDFs, photos of receipts, or scanned faxes. Each adds invisible time. Before accepting an hourly quote, ask what assumptions it rests on: Are the documents already sorted? Is the layout consistent? Do the fields map one-to-one to your system? If the answer to any is no, the hours will climb.

Per-record pricing: false simplicity

A per-record fee looks like certainty until you inspect what "a record" means. Is it one field per record, or ten? Does a record that's 70% missing data still cost the same as one that's complete? What about records that need to be split across multiple lines in your system—does that count as one record or three? Per-record models also assume consistency. If 80% of your data is straightforward but 20% requires research, cross-referencing, or phone calls to verify, the provider's quoted rate may not hold. They'll either absorb the loss or find ways to reclassify those records as "complex" and charge differently. The model also tempts providers to rush: if they're paid per record with no quality gate, speed becomes the priority over care. Before committing, ask the provider to show you examples of how they'd classify and cost a few of your trickier records.

Per-project fees and the margin trap

A flat project fee can feel safe—you know the total upfront. But it transfers the risk of scope creep or changing requirements onto the provider, who then builds a cushion into the quote. That cushion is often invisible to you but visible in the price. A provider quoting per-project has incentives to narrow the scope on paper so the fee looks competitive, then claim extras weren't included once work starts. Conversely, a genuinely fixed-fee quote might exclude anything beyond the core keystroke: no validation against your existing database, no deduplication, no flagging of duplicates for you to merge. You're paying for capture, not for the work that makes captured data useful. With a project fee, get the scope in writing: What constitutes "complete"? Who handles records that don't fit? What happens if you change the field list halfway through? Who validates that the data made it into your system correctly?

What usually gets left out of the quote

The biggest gap between a cheap quote and the true cost is the gap between "data entered" and "data ready to use." A provider can capture text cleanly but miss that a date field is in the wrong format for your system, or that a phone number needs area codes added, or that someone's name is being cut off because the field width is too narrow. These cleanup costs fall to you unless you negotiate them into the service upfront. Quality assurance is another invisible cost—if you're not paying the provider to review their own work, you're paying yourself or another vendor to do it later. Ask whether the quote includes a sample review by the provider before handover and what threshold triggers a re-do. Similarly, clarify whether you're paying for a one-off capture or whether changes requested after handover ("Can you reformat this column?") are included or billable.

The clearest way to compare quotes is not to compare the headline number but to ask each provider the same detailed questions about what they're including, what you're responsible for, and what happens when reality doesn't match the assumptions. Write down their answers. That's when you'll see which quote is genuinely cheap and which is just incomplete.