Competitive Positioning Radar: Read a Competitor's Roadmap From 5 Open Signals

Most competitive intelligence reads like a news wire. "Acme launched a new dashboard." "RivalCo hired a VP of Sales." "CompetitorX raised their enterprise tier 15%." Each observation isolated, stripped of context, delivered weeks after anyone paying attention had already noticed.

The value is not in the signals. It is in the overlap. A hiring spike in healthcare compliance roles. A changelog entry for HIPAA audit logging. A pricing page that just grew a "Regulated Industries" toggle. Read alone, three news items. Read together, a roadmap.

That triangulation says someone is making a deliberate bet on healthcare verticals, probably timed to a Q3 compliance certification announcement. Caught six weeks before the press release, you have runway: accelerate your own healthcare roadmap, reposition before their sales team has new collateral, lock down the at-risk accounts in your pipeline. Caught the day of the announcement, you are reacting.

This article is the system that catches it six weeks earlier.

Job postings leak the most. Before a competitor announces a product, a region, or a segment, they hire for it. Rate of change matters more than absolute count^[3] — a company that jumps from 3 to 12 engineering postings in a single month is signaling something about the next 6-12 months. Not a guarantee. Hiring also reflects attrition or reorg. But the rate-of-change anomaly is where the read starts.

Departmental breakdown sharpens it. An engineering and product spike points to a build phase or platform overhaul. A sales cluster in one geography points to territory expansion. A marketing burst weighted toward demand generation points to a category-creation push. Each pattern is a hypothesis, not a verdict. Combine signals before acting. One channel is noise.

Pricing page changes are the most undermonitored signal. Roughly seven in ten B2B SaaS competitors changed their pricing page at least once over any six-month window^[9] — but most CI programs capture zero of those changes because nobody set up a diff. A competitor who removes their self-serve tier and adds "Contact Sales" is telling you something about their go-to-market motion that no press release will say for another month.

Analyst quotes are the weakest channel, but not worthless. They tell you how a competitor wants to be positioned — their aspirational framing, not their operational bets. Weighted against job postings and changelogs, they help decode intent. Treated as primary signals on their own, they produce the same confident-sounding summaries every news digest generates.

The first version of most radar systems weights all five channels equally. That is a mistake the prediction log will expose within 60 days.

Job postings have the longest lead time and the highest predictive validity for strategic direction. They weight at 1.5x. Changelogs are execution-level signal — they confirm what job postings predicted and narrow the timeline. Weight at 1.2x. Pricing changes are high-conviction signals about go-to-market repositioning, but noisier because they can reflect cost adjustments or A/B tests rather than strategic intent. Weight at 1.2x. Key hires are medium-signal — strong when the hire comes from a domain-specific competitor, weak when it is a generic executive. Weight at 1.0x. Analyst quotes and PR are aspirational signal. They often lag actual product direction by 3-6 months and ghost-written positioning adds no predictive content. Weight at 0.5x.

The first company to calibrate against its own prediction log will tell you these weights are wrong for their specific market and competitor set. That is the point. These are starting weights, not final answers. What stays constant is the principle: channels with longer lead times and harder-to-fake evidence deserve higher weight than channels that report on already-public narrative.

Most teams over-engineer the collection layer. The interesting work is in the inference prompt. The collection layer needs to be reliable and boring — a cron job that runs Monday morning, fetches the last 7 days, and stores results with timestamps and source tags.

For job postings, the practical options are: PredictLeads or Coresignal for structured job data across 2M+ companies (updated every 36 hours)^[3]; direct Greenhouse/Lever/Workday APIs for competitors whose ATS you can identify; or a headless browser hitting the company career page when the ATS is custom. LinkedIn should be treated as a volume indicator (job count from the company page) rather than a scrape target — their ToS and anti-bot defenses have tightened significantly since November 2024, and the IP ban risk exceeds the signal value.

For pricing pages, Visualping at $14/month covers five competitors with twice-daily HTML diffs. The free tier on the Wayback Machine CDX API lets you pull snapshot diffs historically for baseline-setting. If you want to self-host, a weekly playwright screenshot with a pixel-diff against the prior snapshot takes 40 lines of Python and costs nothing.

The system that produces summaries and the system that produces inferences differ by one design choice — the prompt. Most teams ask "what happened?" The right question is "what does this combination of events imply about where this company is heading?" The first generates a digest. The second generates a brief.

An inference prompt that holds up under real signals does three things. It groups signals by competitor and explicitly demands cross-channel convergence. It refuses to output an inference that does not cite a minimum number of supporting signals from different channels. It ties confidence to the diversity and strength of the evidence, not the volume — ten analyst quotes do not equal one hiring spike plus one pricing change.

The first version of this system at one of our portfolio companies weighted all five channels equally. Analyst quotes — largely aspirational, often ghost-written — carried the same weight as job postings. Inference accuracy at medium confidence sat near 40%, well below the 55-65% range we hit after dropping analyst quotes to a 0.5x multiplier and lifting job postings to 1.5x. The lesson is structural: the channels are not equally predictive, and your initial weights are almost certainly wrong. Calibrate against your own prediction log or accept that the radar is decoration.

The second most common failure is the "what would increase confidence" field going missing. That field is not decorative — it produces your monitoring agenda for the following week and prevents the inference from becoming a closed narrative.

The downmarket move is the hardest to read because it often shows up as a reduction in signals rather than a spike. Engineering headcount holds steady while a product-led growth hire appears. The changelog starts showing simplified onboarding flows. A new pricing tier appears with a lower floor or a free plan. Without a baseline, you miss the relative shift. This is why baselines are mandatory before you run the first inference cycle — the absence of a spike is sometimes the signal.

A competitor going through a leadership transition leaks differently. The VP Engineering departure followed by a 60-day hiring pause is not a strategy signal — it is an execution signal that tells you the product roadmap just got reprioritized. That is worth knowing too: it means the next 90 days will see slower delivery and potentially a window for your sales team to reopen deals they lost to that competitor.

Every intelligence program is tempted to measure output volume — briefs published, signals collected, competitors tracked. These metrics tell you the machine is running. They tell you nothing about whether the radar changes a decision.

Three metrics are load-bearing. Prediction accuracy over 90 days: what percent of medium-and-high confidence inferences proved correct against actual outcomes? Practitioners report clusters around 55-65% accuracy at medium confidence and 70-80% at high confidence when the channel weights are calibrated^[1]. Below those bands, the signal collection or the inference prompt needs recalibration — usually the prompt.

Time-to-action: when a high-confidence inference is published, how many days pass before a stakeholder takes a measurable action — adjusts a roadmap, modifies positioning, reaches into an at-risk account? Briefs sitting unread for two weeks are a distribution problem, not an intelligence problem. Fix the format and the recipient list before touching the model. Crayon's 2025 research found that teams enabling sales daily with CI saw an 84% increase in competitive sales effectiveness — and that number moved further when executive sponsorship was added^[10].

Competitive win-rate delta: across a rolling quarter, compare win rates on deals where the team had advance intelligence from the radar against deals where they did not. A working program produces a measurable improvement. When a rep received relevant battlecard intel within 27 minutes of a competitor being mentioned on a discovery call, win rates jumped from 32% to 67% — a concrete example of what time-to-action actually means at the deal level^[10]. Win-rate delta is the metric that ends the budget conversation.

One thing most teams never measure: false positive cost. Every high-confidence inference that turns out wrong consumes leadership attention and, if acted on, resources. Tracking your false positive rate by channel reveals which channels are dragging accuracy down. The fix is almost always the same — reduce the weight of the noisiest channel.

The fastest working radar is set up in a week. Pick the two competitors your sales team encounters most often. Map their career page, changelog, and pricing page. Set up weekly diff monitoring on each URL — Visualping at $14/month covers five URLs with twice-daily diffs, or a cron-driven headless browser snapshot if you want self-hosted.

Write one inference prompt using the template in this article. Feed it the first week's signals. The output will be flawed. It will also be more useful than a Slack channel full of "hey, did you see CompetitorX launched a new feature?" messages.

Four weeks of data is enough to set meaningful anomaly thresholds. Eight weeks is enough to see prediction accuracy stabilize. Twelve weeks is enough that the team stops asking why the radar exists — and starts asking why it wasn't built sooner.

The Crayon 2025 data point worth holding onto: the average CI program takes 6-9 months to show statistically significant win rate improvement^[10]. That doesn't mean the radar has no value before then. The value in weeks 1-12 is decision speed — catching a competitive move before the press release and having time to respond. The win rate delta shows up later. Don't let the absence of that metric early on kill the program before it has time to prove itself.

The edge in competitive markets is not better products. It is shorter latency between the competitor's move and your response. The signals have always been there. The difference is whether you have a system that reads them, weights them correctly, and turns them into a decision — not a summary.