The outside-in questions
Every market has a few facts that shape every decision made inside it, and a business that does not know them is planning blind. How many people or organisations could buy what we sell, and how many realistically will? Is that number growing, flat or shrinking, and why? What is changing around the market: regulation, technology, costs, the way people buy? Who holds the demand today, what do they offer and at what price, and what are they about to do? Which gaps exist between what buyers want and what the current competitors provide?
These are questions about the world rather than about any one customer, and most of them have been studied by someone: official statistics offices, regulators, trade associations, analysts, the financial reports of listed competitors, and the search engines whose query data records what people are looking for. The work is finding, judging and combining those sources, and filling the remaining gaps with primary research only where nothing else will serve. The marketing research hub places this outside-in work within the wider process; this section covers what is specific to it.
Sizing, trends, competitors
Three kinds of outside-in question recur, each with its own methods and its own traps.
| Question | What it asks | Main methods | Main trap |
|---|---|---|---|
| Sizing | How large is the demand we could serve, and the share we could win? | Top-down from published figures; bottom-up from counts of buyers and purchase behaviour; proxies such as search demand | A single large figure from a report, divided by a hopeful share, presented without a range |
| Trends | Which way is the market moving and what is driving it? | Time series from official and trade sources; search interest over time; regulator and policy reading; a PESTEL pass | Mistaking a short run of data for a direction; reading a vendor's forecast as a fact |
| Competitors | Who competes for the same demand, how, and what are they changing? | A one-off structured analysis; then ongoing monitoring of positioning, pricing, messaging, hiring and media activity | Analysing once, never updating; watching the companies the founders fear rather than the ones customers weigh |
Sizing and competitors have their own guides in this section. Trend research threads through both: a size without a direction is a snapshot, and a competitor's moves only make sense against what the market is doing.
The core guides
- Market sizing: TAM, SAM and SOM — how to define total, serviceable and obtainable markets so the definitions survive questioning, how to build top-down and bottom-up estimates and make them agree, which proxies (search demand, category data, trade sources) are worth using, and how to present a range honestly.
- Competitive intelligence — the ongoing programme that follows a one-off analysis: the sources from ad libraries to job adverts and what each reveals, how to track changes in positioning, pricing and messaging, share of voice and search visibility as signals, a monitoring cadence, and the ethical and legal limits.
The one-off competitor analysis itself, the structured pass that defines the competitive set and compares positioning, offer and channels at a point in time, belongs to the strategy section and is covered by the competitor analysis guide. Read that first; the intelligence guide here assumes it has been done and covers what to watch afterwards.
Further guides
Further guides cover the secondary pass that comes before any fieldwork, the people who know a market without being its buyers, and the share of the market the business already holds:
- Running a desk research pass — a tutorial for turning the decision into questions, searching internal data and published sources, logging and judging each source, and reconciling the figures.
- Interviewing industry experts and the trade — who knows the market, what to ask, the lines on competitor information, and weighing what they say.
- Estimating your market share — share by value or volume on the sizing definition, penetration beside it, routes to an estimate, signals that move with share, and a dated range read over time.
Secondary sources and how to judge them
Most market research is reading, and the quality of the research is set by how the reading is judged. Sources fall into rough tiers. Official statistics and regulator publications (in the UK, for instance, the Competition and Markets Authority publishes market studies and investigation findings on sectors it has examined) have documented methods and no commercial interest in the answer, but they can lag the market and define categories differently from how a business sees them. Trade associations know their industry well and have an interest in it looking healthy. Analyst reports vary enormously in rigour and are often priced beyond small teams. Vendor-published research exists to generate leads, and its numbers are chosen to make a point. Competitors' own public statements, annual reports and investor materials are accurate about what the company must disclose and promotional about everything else.
For any source, ask five questions. Who produced it and what did they want from it? How was the data collected, from whom and when? How does the source define the market, and does that match ours? Can the figure be reconciled with a second, independent source? And what would the figure look like if the source's interest ran the other way? A number that survives those questions carries weight; a number that cannot be traced to a method is a rumour with a decimal point. The desk research tutorial puts these questions into a pass: log every source before using it, record its weight, extract figures with their definitions and dates, and reconcile the ones that disagree. Some knowledge sits with people rather than in publications, and the guide to interviewing industry experts covers distributors, installers, consultants and trade journalists, and how to weigh what they say against published figures and customer evidence.
Search data is a source of its own kind. Query volumes and the composition of a SERPThe page of results a search engine shows. show what people are looking for and who is answering, and they are current in a way published reports are not. They are also partial: search records expressed demand from people who search, which is not all demand and not all people. Used as a proxy alongside other sources rather than as a market size in itself, search data is one of the most useful inputs a marketing team has.
Turning market research into strategy inputs
Market and competitor findings are not a report; they are inputs to decisions that live elsewhere. The size and direction of the market feed the choice of which segments to serve and whether to enter at all. The competitive picture feeds positioning, because a position is a claim relative to alternatives, and pricing, because price is read against what competitors charge. The forces acting on the market feed planning assumptions and risk registersThe single table of what could knock a marketing plan off course, each risk with a warning sign, an owner and a response prepared in advance.. A SWOT built without this research is a list of opinions; built on it, the external half (opportunities and threats) has evidence behind each entry.
Two measures carry the outside-in view into the regular reporting. Share of VoiceYour share of market visibility., the business's share of the category's advertising or visibility against competitors, is a leading signal of market share movement, and tracking it is part of the intelligence routine. Market share itself, where it can be estimated, is the lagging KPIA metric tied directly to an objective. that tells the business whether the strategy is working. Both are only meaningful if the market they are measured against is defined the same way the sizing work defined it, which is one more reason to write the definitions down.
A software firm reads its market before a launch
Tallyspire Compliance, a small vendor of audit software for food manufacturers, is deciding whether to launch into the UK drinks sector. The marketing lead begins with secondary sources: the official business population statistics put the number of drinks manufacturers at roughly 2,400, a trade body report suggests around a third already use specialist compliance software, and search data shows a few hundred monthly queries for the category's core terms. She sets the serviceable market at firms with more than ten employees, about 900, and checks the figure against the customer counts two listed competitors disclose in their annual reports, which together reach roughly 260 drinks manufacturers. Her bottom-up estimate of the obtainable market in three years is 60 to 90 customers. The competitor review finds both rivals pricing per site rather than per user, and a six-week scan of their job adverts shows one hiring a drinks-sector sales specialist. The launch goes ahead, priced per site, with a positioning aimed at the mid-sized firms the hiring suggests the rival is also targeting.
Common market research problems
| Problem | Usual cause | Guide |
|---|---|---|
| A market size nobody will defend in front of the board | Top-down figure with no bottom-up check, no stated definition and no range | Market sizing |
| Serviceable and obtainable markets blur into the total | Definitions never written down, or written to make the number look large | Market sizing |
| The team learned of a competitor's repositioning from a customer | No monitoring routine after the one-off analysis | Competitive intelligence |
| Competitor monitoring produces noise, not decisions | No cadence, no owner, no routing of findings to the people who act on them | Competitive intelligence |
| The competitor set is the one the founders fear, not the one buyers weigh | Competitive set defined from inside the business rather than from customer research | Competitor analysis |
| A vendor's forecast became the plan's growth assumption | Sources not judged for interest and method | Market sizing, Running a desk research pass |
| Share of voice or share of search is reported with no market share beside it | No estimate of the business's current share on the sizing definition | Estimating your market share |
| One expert's view became a figure in the plan | Expert opinion recorded without its basis and never weighed against published figures or customer evidence | Interviewing industry experts |
| Intelligence gathering made the legal team nervous | Methods that cross from public observation into misrepresentation or improper information exchange | Competitive intelligence, Interviewing industry experts |
