A step by step competitive analysis you can run in one week

Run these seven steps in order and you have a working competitive analysis: define your decision and scope, map every direct, indirect and substitute competitor, collect primary and secondary data against a fixed template, build a comparison matrix across features and pricing, synthesise the findings into a SWOT and TOWS, prioritise the resulting actions with a scoring rubric, then set a monitoring cadence so the whole thing doesn't go stale in six weeks.

A quick audit covering a small number of rivals takes one person a few working days. A full competitive analysis covering several competitors, with primary research and a stakeholder presentation, typically runs a small team several working weeks. Shopify's own competitive analysis playbook backs the seven to ten competitor range as the sweet spot: wide enough to catch blind spots, narrow enough that nobody drowns in spreadsheets.

The finished deliverable should be boring in the best sense: a one-page executive summary stating the recommendation, backed by a slide deck or spreadsheet holding the comparison matrix, SWOT/TOWS grid, and a prioritised action list with named owners. If your output can't fit on one page, you haven't finished the synthesis step yet.

1. Scope — pick one business decision this analysis will inform.
2. Map competitors — direct, indirect, and substitutes.
3. Collect data — primary interviews plus secondary sources.
4. Compare — build the feature, pricing, and positioning matrix.
5. Synthesise — turn the matrix into SWOT/TOWS.
6. Prioritise — score actions on impact versus effort.
7. Monitor — set a cadence and assign an owner.

Table of Contents

• How do you carry out a step by step competitive analysis?
• How do you identify direct, indirect and substitute competitors?
• Where do you find reliable market research and data?
• What should you compare when analysing competitor offerings?
• How do you turn competitor findings into a strategy?
• What templates make this process repeatable?
• How long does a competitive analysis take and what does it cost?
• What mistakes and red flags undermine competitive analysis?
• How can AI signal detection strengthen ongoing monitoring?
• Key takeaways: the essential checklist and a 48-hour plan
• Sources

How do you carry out a step by step competitive analysis?

Each step below has a checklist, an acceptance test, and a suggested owner, because a step without an owner is a step that never happens.

Step 1: Define objectives and scope. Start by naming the actual decision this analysis needs to enable, whether that's a pricing change, a new market entry, or a product roadmap call. Vague scope is the single biggest reason competitive analyses get built once and never opened again. The Ahrefs guide to lean competitive analysis makes the case bluntly: gather only the data that answers your decision, not everything that looks interesting.

• Checklist: write the decision in one sentence, name the deadline, list who signs off on the recommendation.
• Acceptance criteria: a stakeholder can read your scope statement and say exactly what decision it will change.
• Owner: the project lead or marketing strategist commissioning the work.
• Deliverable: a one-paragraph scope brief.

Step 2: Map the competitive landscape. At this stage you're building the candidate list, not analysing anyone yet. Cast a slightly wider net than feels comfortable, because the competitor that beats you in three years rarely looks like a competitor today.

• Checklist: list direct rivals, indirect rivals, and substitutes; note who came up in the last five customer conversations.
• Acceptance criteria: at least seven to ten named competitors, each with a one-line reason for inclusion.
• Owner: analyst or junior marketer.
• Deliverable: a candidate competitor list.

Step 3: Collect the right data. Split effort between primary sources (customer interviews, mystery shopping, product trials) and secondary sources (industry reports, government statistics, traffic tools). Log where every fact came from and when you pulled it, because a data point without a date is a liability, not an asset.

• Checklist: assign each competitor a data owner; log source and date for every entry.
• Acceptance criteria: no comparison cell without a source citation.
• Owner: research analyst, split by competitor or data category.
• Deliverable: a source log alongside the raw data table.

Step 4: Build comparison artefacts. Turn raw notes into a feature matrix, a pricing grid, and a positioning map. This is where a scrappy spreadsheet becomes something a chief marketing officer can actually read in ninety seconds.

Step 5: Synthesise findings into SWOT and action options. Every strength or weakness in your SWOT should trace back to a specific row in your comparison matrix. If it doesn't, it's an opinion, not a finding.

Step 6: Prioritise opportunities. Score each option against impact, ease of execution, and strategic fit, then assign named owners and deadlines.

• Checklist: score every action 1 to 5 on impact and 1 to 5 on ease.
• Acceptance criteria: every action above the priority threshold has an owner and a deadline.
• Owner: department head or strategy lead.
• Deliverable: a ranked action register.

Step 7: Set up continuous monitoring. A one-off report goes stale the moment a competitor changes their pricing page. Building monitoring cadence and ownership into the process from day one converts a static document into what Monday.com's guide to continuous competitive intelligence calls a decision system rather than a shelf report.

Pro Tip: Timestamp every data point the moment you collect it. Six weeks later, "traffic was up" means nothing without a date attached to it.

How do you identify direct, indirect and substitute competitors?

Getting the competitor set right matters more than any spreadsheet formula that comes after it. Miss the wrong company and your entire matrix is comparing yourself to the wrong bar.

Direct competitors sell the same solution to the same buyer. Indirect competitors solve the same underlying problem with a different approach, think a scheduling app competing against a virtual assistant service. Substitutes are what a customer does when they give up on buying anything at all, like a spreadsheet standing in for a subscription tool. The U.S. Small Business Administration's guidance on competitive analysis recommends identifying competitors by product line or market segment first, then assessing products, pricing, distribution, and promotion for each one.

Practical search techniques beat guesswork here. Run keyword overlap searches to see who else ranks for your core terms. Search the marketplaces your customers actually browse, whether that's an app store, a B2B directory, or Amazon. Monitor paid advertising on the keywords tied to your product category, since ad spend is one of the more honest signals of who considers you a threat. Map the entire customer journey and note every alternative a buyer might encounter at each stage, not just the ones you already know by name.

Prioritise your list using three questions: does this competitor affect the decision you scoped in step one, do they show up repeatedly when customers compare options, and do they have visible share of voice in your category's search results or review sites? A competitor who ticks all three deserves a full profile; one who ticks none can sit in a watchlist instead.

For your candidate list, capture these fields for each entry: company name, competitor type (direct, indirect, substitute), primary product, estimated market position, and the specific reason they made the cut. Seven to ten well-chosen names beats twenty half-researched ones every time.

Where do you find reliable market research and data?

Good competitive analysis rests on data you can defend in a meeting, not screenshots pulled from a competitor's marketing page the morning of the presentation.

Collect data across eight categories: market size and growth, pricing, traffic and share of voice, customer sentiment, product features, distribution channels, and hiring or operational signals (job postings are an underrated tell for where a competitor is investing). For market scoping, the Census Bureau's NAICS classification system gives you a standardised way to define your industry boundaries, which matters more than it sounds when two companies dispute whether they're even in the same market.

For broader economic context, the Bureau of Economic Analysis's consumer spending data is a genuinely authoritative source for demand trends, and the Bureau of Labor Statistics' Consumer Price Index gives you a benchmark for whether a competitor's pricing move reflects real cost pressure or pure margin grab. Census Business Builder adds county-level business pattern data if you need to quantify how many competitors actually operate in a given geography. Commercial traffic tools like SimilarWeb and Google Trends fill in the digital-visibility gaps that government data doesn't cover.

Primary research still beats secondary data for anything qualitative. Customer interviews, mystery shopping a competitor's sales process, running a free trial of their product, and testing their purchase funnel end to end all surface details no report will ever mention, like how long their checkout takes or what their support team actually says when pressed.

Before you trust any data point, run three validation checks: does the source disclose its methodology, is the data recent enough to matter for your decision, and does it agree with at least one independent source? A single unverified claim from a competitor's own press release doesn't clear that bar on its own.

Pro Tip: Government sources rarely tell you about a specific competitor, but they're the fastest way to sanity-check whether a rival's growth claim is plausible against the actual market's size.

What should you compare when analysing competitor offerings?

A comparison matrix is only as trustworthy as the dimensions you choose to fill it with. Skip a dimension and you get a tidy table that misses the reason customers actually switch.

Build your matrix across six dimensions: product and feature set, pricing and monetisation model, target customer segment, go-to-market channels and messaging, strengths and unique selling points, and customer experience including support and adoption signals. The SBA's guidance on assessing "products, pricing, distribution and promotion" maps almost exactly onto this structure, and it's held up well because those four areas are genuinely where most competitive advantage lives.

Populate the matrix using a simple evidence tag next to every entry: primary (you tested it yourself), secondary verified (a credible third-party source), or secondary unverified (a claim you haven't confirmed). This single habit stops a competitor's own marketing copy from quietly becoming "fact" in your final report.

1. Score each competitor 1 to 5 on every dimension.
2. Weight dimensions by relevance to your scoped decision, not evenly.
3. Normalise scores across competitors before comparing totals.

For visualisation, a feature heatmap works well for spotting gaps at a glance, a positioning map (price against a second axis like feature breadth) clarifies where genuine white space sits, and a ranked scorecard turns the whole exercise into a single number executives can argue about. Coursera's step-by-step guide to competitor analysis recommends exactly this combination: share of voice, pricing and feature comparison, and review analysis, run together rather than in isolation.

Translate scores into a recommendation by asking one question of the finished matrix: where is the gap between what customers want and what every competitor currently offers? That gap is your opening.

How do you turn competitor findings into a strategy?

Raw comparison data doesn't make decisions by itself. SWOT and TOWS are where the matrix becomes an argument for doing something specific.

Build your SWOT using nothing but evidence already sitting in your comparison matrix. A "weakness" isn't a weakness unless a row in your data supports it; otherwise it's an assumption wearing a strategy hat. Once you have Strengths, Weaknesses, Opportunities, and Threats laid out, the TOWS matrix forces the next step: pairing each Opportunity with a Strength to generate an offensive move, and pairing each Threat with a Weakness to generate a defensive one.

• Strength + Opportunity: use a proven capability to chase an underserved segment a competitor has ignored.
• Weakness + Threat: shore up a known gap before a well-resourced rival exploits it.
• Strength + Threat: lean on an existing advantage to blunt a competitor's aggressive move.
• Weakness + Opportunity: decide whether closing a gap is worth the resource before chasing new demand.

Score every resulting action on three axes: impact (how much this moves the needle on your original decision), ease (resourcing and time required), and strategic fit (does this align with where the business is already heading). A high-impact, low-ease action isn't automatically wrong, but it needs a longer runway and a more senior sponsor than a quick win does.

Every recommended action needs acceptance criteria before it goes on the priority list, meaning a measurable outcome and a deadline, not just a good idea. Attach a KPI to each one: a pricing change might target a specific conversion lift, a positioning shift might target a change in branded search volume, a support investment might target a reduction in churn among a named segment.

Pro Tip: If a SWOT item can't be traced back to a specific row in your comparison matrix, delete it. Untraceable insights are the fastest way to lose a stakeholder's trust in the whole report.

What templates make this process repeatable?

Reusable assets are the difference between a one-off exercise and a method your team actually runs again next quarter. Shopify's own research on competitive playbooks found that teams are far more likely to reuse the output when it comes packaged as a template plus a worked example rather than a blank framework.

• A spreadsheet with tabs for the competitor list, the data collection log, the comparison matrix, and the SWOT/TOWS grid.
• A slide deck outline: one slide for the decision being enabled, one for the top three findings, one for the matrix summary, one for prioritised actions with owners.
• A customer interview script with five to seven open questions covering how buyers currently solve the problem and what would make them switch.
• A source log template capturing competitor, data point, source, date, and confidence level for every entry.

Here's a compressed worked example. Say the decision is whether to introduce a mid-tier pricing plan. Step 2 maps four direct competitors and two substitutes. Step 3 collects pricing pages, three customer interviews, and BLS price-index context to check whether the market is under general inflationary pressure. Step 5's SWOT flags this gap as a Weakness against an Opportunity (unmet demand from customers who find the entry plan too limited).

Adapt the same templates for other decisions by changing what step 1 scopes. A market entry decision needs an extra column for regulatory or logistical barriers; a product roadmap decision needs a heavier weighting on the feature matrix relative to pricing.

How long does a competitive analysis take and what does it cost?

Time and cost scale directly with how many competitors you cover and how much of the data collection is primary research versus desk research.

A quick audit of three to five direct competitors, using only secondary sources, takes one analyst two to three working days. A full competitive analysis covering seven to ten competitors, including at least a handful of customer interviews or mystery-shopping exercises, runs ten to fifteen working days for a small team of two or three people spread across research, analysis, and presentation roles. An ongoing monitoring programme typically needs two to four hours a week from a named owner, plus a monthly half-day review to update the matrix and flag anything material.

Minimum team roles look like this: one research analyst for data collection, one strategist or marketer for synthesis and prioritisation, and one stakeholder sponsor who signs off on the final recommendation. For a quick audit, one person can cover all three roles.

On outsourcing: keep synthesis and prioritisation in-house, because nobody outside the business understands your strategic constraints well enough to weight the TOWS matrix correctly. Data collection, particularly desk research against government and industry sources, is easier to hand off to a junior analyst or contractor without losing quality.

Tools worth budgeting for include a traffic and SOV tool such as SimilarWeb, a review-aggregation tool for sentiment analysis, and a shared spreadsheet or lightweight BI tool for the comparison matrix. None of these require enterprise licensing for a first pass.

Pro Tip: If your budget only stretches to one paid tool, put it toward traffic and share-of-voice data. Pricing and feature data you can usually collect by hand; digital visibility trends are much harder to estimate manually.

What mistakes and red flags undermine competitive analysis?

Most competitive analyses fail quietly, not because the research was wrong but because of how it was collected and interpreted.

• Over-collection: gathering every data point available instead of only what answers the scoped decision, which produces a "scrapbook" nobody opens twice.
• Confirmation bias: cherry-picking evidence that supports what leadership already believes about a competitor.
• Mistaking marketing claims for evidence: treating a competitor's press release or landing page copy as fact without independent verification.
• Small-sample inference: drawing conclusions about an entire market from two or three customer interviews.
• Stale data: presenting pricing or feature information that's months old as if it reflects the current state of play.

Red flags in the data itself deserve extra scrutiny. Inconsistent reporting between a competitor's own investor materials and their marketing claims usually means one of the two is inflated. A sudden traffic spike with no accompanying product launch, press mention, or seasonal explanation often signals a paid campaign or a data anomaly rather than organic growth. Reviews that arrive in unnatural clusters, all five stars, all posted within days of each other, are a classic sign of manufactured social proof rather than genuine customer sentiment.

Before sharing results, run a short quality-control pass: check that every claim in the executive summary has a source with a date, confirm no single unverified data point carries the whole recommendation, and have someone outside the research team read the SWOT and ask "how do you know that?" for each item.

How can AI signal detection strengthen ongoing monitoring?

Manual monitoring catches what you already know to look for. It's much weaker at catching the thing nobody on the team has thought to search for yet.

An early signal typically starts as noise: a small spike in job postings for a niche skill, a cluster of forum mentions using new terminology, a sudden shift in ad creative themes across a category. Most of these signals never turn into anything. The ones that do usually show up months before they're obvious in a competitor's official announcements or pricing pages, which is the entire value of catching them early rather than reading about them in a trade publication after the fact.

Validating a signal properly means triangulating it against at least two independent sources and having a human review it before it changes a decision. A single noisy data point, one unusual traffic spike, one odd job listing, is not a trend. Three independent signals pointing the same direction across different data types (hiring, search behaviour, and product changes, say) is a much stronger basis for action.

Integrate signal monitoring into the same cadence you set up in step seven of the core process. Assign an owner, decide how often the signal feed gets reviewed against your existing comparison matrix, and set a threshold for when a signal is strong enough to trigger a full re-analysis rather than a watch-and-wait note. OnTheRice's own approach uses multiple AI engines to extract early signals from noisy global data and produce ranked scores, which is the kind of automated first pass that can meaningfully cut the time between "something is shifting" and "we've validated it and built a response."

Pro Tip: Treat any single AI-flagged signal as a hypothesis, not a conclusion. The value is in surfacing it faster than a manual scan would, not in replacing the human triangulation step.

Using competitor insight to set quarterly priorities

Most teams treat competitive analysis as a one-time report handed up the chain and quietly forgotten. That's the wrong model. The genuinely useful version treats the matrix as a living input to quarterly planning, reviewed and re-scored every time priorities get set, not just when someone remembers it exists.

Small, evidence-backed changes tend to outperform sweeping strategic pivots built on the same research. A pricing tier adjustment informed by a genuine gap in the matrix, or a support-response commitment built off a documented weakness in a competitor's customer experience, moves faster and carries less risk than a full repositioning built on the same data. The discipline is in matching the size of the action to the strength of the evidence behind it.

On governance, the analysis only stays useful if someone owns the cadence. That means a named person checking the monitoring feed weekly, a monthly session where the comparison matrix gets updated, and a quarterly review where the SWOT and TOWS get rebuilt with fresh evidence rather than recycled from last time. Stakeholder engagement matters just as much as the research itself: bring the matrix into planning meetings as a working document people can challenge, not a finished report people are asked to approve.

Automated signals and ranked trend cards from OnTheRice

Ontherice is the practical answer to step seven of this whole process, the monitoring cadence that most teams set up with good intentions and then let lapse within a month. Instead of manually re-checking pricing pages and job boards, you get AI-generated opportunity signal cards that scan public data continuously and rank what's actually gaining momentum before it's obvious to everyone else watching the same market.

The platform maps directly onto the monitoring checklist covered above: emerging brand tracking for competitor movement, transparent scoring so you can see why something ranked where it did, and live AI queries when you need a fast answer instead of a full re-run of the matrix. If you're tracking who's rising in your category before their next funding announcement or product launch makes it obvious, explore emerging brand rankings and set up your first monitoring feed today.

Key takeaways: the essential checklist and a 48-hour plan

A competitive analysis only creates value when it's scoped to one decision, built on dated and sourced evidence, and paired with an owner who keeps monitoring it after the report is delivered.

1. Write your scope statement today: name the decision, the deadline, and who signs off.
2. Build your candidate competitor list within 24 hours, aiming for seven to ten names across direct, indirect, and substitute categories.
3. Pull your first three secondary data sources within 48 hours, prioritising government or industry sources you can trust without verification.
4. Draft the comparison matrix skeleton before your first stakeholder check-in, even if only a few rows are filled in.
5. Assign a monitoring owner in the same meeting where you present findings, not as an afterthought later.

Sources

The sources below cover the two data problems every competitive analysis eventually hits: quantifying the overall market, and spotting shifts before they become obvious.

• Market research and competitive analysis | U.S. Small Business Administration
• NAICS - U.S. Census Bureau
• Consumer Price Index (CPI) - Bureau of Labor Statistics
• Consumer Spending - Bureau of Economic Analysis
• How to perform competitor analysis: A step‑by‑step guide | Coursera
• Competitive Analysis Guide 2026: Free Template + AI Tools - Shopify

Recommended

• Competitive insight checklist for business strategists
• Gain strategic advantage with competitive trend monitoring
• Industry ranking checklist for UK market strategists
• Market insights software comparison: AI-native vs augmented