Part of our series: The People Behind the Research
Statpit publishes nearly 10,000 industry reports across 120 industries, alongside software Best Lists with a numbers-first focus on pricing transparency and total cost of ownership.
---
Adrien, you fact-check a large share of Statpit's Best Lists, and your own beat is luxury goods. That's an unusual combination.
Adrien Chevalier: It's a useful one. I have a Master's in Marketing Analytics and a Bachelor's in Business Administration, and I spent six years as a luxury goods researcher at an independent luxury market advisory firm in Copenhagen. At Statpit I cover luxury goods economics, premium retail and brand valuation. Luxury teaches you that price is never just a number. It's a positioning decision, and it's often deliberately hard to compare. Brand valuation is similar: you can get very different answers depending on the method, so the method always has to sit next to the figure. When I fact-check a Best List, I apply the same rule. If a page states a price, I want the source, the date it was checked and the tier it refers to. Software pricing changes constantly, and a price without a date is a rumour.
Magnus, you write many of the Best Lists. What's your background?
Magnus Öberg: Operations research and mathematics: a Master's in Operations Research and a Bachelor's in Mathematics. I spent four years as a logistics researcher at an independent supply chain advisory firm in London. At Statpit I cover last-mile delivery, warehouse automation and transportation management. Operations research is essentially the discipline of modelling costs honestly, including the ones that only appear at scale. Logistics software is a good example. It's often priced per shipment, per route, per location or per user, and the cheapest option at your current volume can be the most expensive one after a year of growth. That's the thinking I bring to every list, whatever the category. The price at ten users is not the price at five hundred.
Eero, you cover data engineering and data warehousing, probably the hardest pricing in software to compare.
Eero Mattsson: It really is. I have a Master's in Computer Science and a Bachelor's in Statistics, and I spent six years as a data engineering researcher at an independent analytics technology advisory firm in Milan before freelancing as a data engineering analyst. Data platforms increasingly price on consumption: compute, storage, credits, data transfer. That means there is no single price. There is a function of your workload. Two vendors can publish rates that look comparable and produce very different bills because they meter different things. So when I compare them, I state the workload assumptions explicitly. If a reader's workload differs, they can see exactly where the comparison stops applying to them.
Carina, you're on the statistics side of Statpit. How do you approach it?
Carina Vogel: As a statistician first. I hold a Master's in Statistics and a Bachelor's in Sociology, and I spent six years as a demographic researcher at an independent population studies advisory firm in New York before freelancing as a demographic analyst. At Statpit I report on demographic trends, migration statistics and urbanization data. Official statistics look authoritative, and mostly they are, but they come with definitions, revisions and vintages that get lost when numbers travel. "Migrant" means different things in different national statistics. Population estimates get revised. A figure quoted without its year and its definition can be technically correct and still wrong for the comparison someone is making. My job is to keep those attachments on the number.
---
Statpit calls itself "numbers-first." What does that look like in your process?
Carina: Three stages. First, primary-source research: analysts pull figures from official statistics, filings and established databases, and humans decide what enters the pipeline. Second, cross-tabulation and checks: we compare numbers across sources, run automated and cross-model checks, and flag gaps. AI helps with retrieval and extraction at a speed no team could match manually, but it's a tool in the pipeline, not the editor. Every line should point to a traceable reference before it goes live. Third, an editor reviews sourcing and consistency and makes the final call.
Adrien: And we label how much support each figure carried through review. The target mix across row-level indicators is about 70 percent verified, 15 percent directional and 15 percent single source. It's a transparency signal, not a legal warranty, and we say so. But it means a reader never has to guess whether a number was confirmed by three sources or by one.
Magnus: Cross-tabulation is where most of the problems show up. When two credible sources disagree, that's not something to smooth over. It usually means they are measuring different things, and figuring out which is the actual work.
How do you research software pricing when so many vendors don't publish it?
Magnus: We look at three layers: list prices, tier logic and scaling costs. List prices are the published numbers. Tier logic is what you actually have to buy to get the feature you need; the classic case is a single capability that's only available two tiers up. Scaling costs are what happens as seats, volume or usage grow.
Eero: And when pricing is opaque, we say so plainly. "Contact sales" is a data point. It tells a buyer they will be negotiating without a public reference, which matters for budgeting and for leverage. We'd never invent an estimate to fill the gap, but we won't hide the gap either.
Adrien: Tools with public pricing can be compared honestly. Tools without it can still be good tools, but the reader should know they can't verify the cost in advance.
---
Some of your studies are funded by software vendors. How does that work?
Adrien: Vendors fund studies because they want rigorous, comparable research on their market to exist, and that research is expensive. So each study is co-funded by several of the vendors it assesses. Sharing the cost is what keeps rigorous research affordable, and it means no single vendor ever funds a study.
Carina: Funders may propose the market to be studied. They don't choose which vendors are included, they don't set or weight the criteria, and they don't approve findings. Before research begins, an editor fixes the market definition, the inclusion rules and the evaluation criteria with their weights. That protocol is published with the study and doesn't change once research starts. From a statistician's point of view, that's the most important safeguard: the analysis plan exists before the data does.
Eero: Every vendor in scope, funder or not, is assessed against the same criteria and offered the same factual-accuracy check on its own products before publication. Corrections are made only where documentation supports them, and nobody gets to review what we say about other vendors.
Magnus: Every completed study is published as found. There is no option to fund a study and then shelve it because the results were inconvenient. Each one is labelled as vendor-funded at the top of the page and carries a funding statement in its methodology. And studies are kept separate from everything else: they are never cited as evidence in our statistics reports or Best Lists, and they have no effect on rankings.
---
What's the most misleading number in your field?
Eero: The headline per-unit rate. It's real, but it's rarely what drives the bill.
Magnus: "Starting at." It describes the cheapest configuration that almost nobody buys.
Carina: A population figure without a year. Revisions can move numbers more than people expect.
Adrien: Brand value presented as a single objective figure. There are several established valuation methods, and they don't agree. Pick one and say which.
Last question: what should readers know about Statpit that isn't obvious from the site?
Magnus: That we model the second year, not just the first month.
Eero: That stating our assumptions isn't a disclaimer. It's the most useful part of the comparison.
Carina: That a number without its definition and its date isn't finished, and we don't publish unfinished numbers.
Adrien: That every price on a Best List was checked by someone who cared which tier it belonged to.
---
Statpit.com publishes nearly 10,000 free research reports and software Best Lists across 120 industries. Explore the library at statpit.com/topics/ and read the full methodology at statpit.com/editorial-process/.