Key finding:
- Voxagent has no measurable search presence: 0 organic keywords, 0 organic traffic, DR 0, and zero AI visibility across ChatGPT, AI Overviews and AI Mode.
- All four competitors are established in Russian search — Zvonobot (22.8K visits/mo, DR 51), Skorozvon (17.3K, DR 39), Twin24 (11.3K, DR 59) and Voxys (2.3K, DR 42).
- The addressable Russian-language demand is 29.2K searches/month across 9 clusters — a genuinely niche market: the head term голосовой робот draws only ~500 searches/mo, so the strategy must win on breadth and intent depth, not on head-term volume.
- Only 370 referring domains and 380 backlinks — essentially no authority base to rank from yet.
Content Opportunities
241
Relevant content ideas (excl. off-topic)
9
Relevant topical clusters
29,240
Relevant monthly search volume
| Content Cluster | Content Ideas | Monthly Search Volume |
|---|
| AI Agents & Assistants for Business | 49 | 7,570 |
| Voice Robots & Bots (core) | 44 | 4,290 |
| Outbound Calling & Auto-dialling | 33 | 4,010 |
| Call Centre Operations | 36 | 3,560 |
| Speech Analytics | 30 | 3,260 |
| Business Telephony & Cloud PBX | 14 | 2,870 |
| Sales & CRM Automation | 20 | 2,130 |
| Voice Robots for Inbound Calls | 14 | 1,540 |
| Pricing & Vendor Selection | 1 | 10 |
| Off-topic / adjacent tools — excluded | 67 | 32,450 |
| Total (all) | 308 | 61,690 |
How this was built
- Seed keywords derived from voxagent.ru site content, expanded via Google Ads Keyword Planner (308 keyword ideas) for the client's live traffic geographies.
- Clustered by search intent; every keyword sits in exactly one cluster.
- The off-topic row is excluded from the three headline figures — it holds adjacent-tool and generic queries the Planner returns but that carry no purchase intent for this product.
- 👆 Click any cluster row to see its top keywords by volume.
Competitive Landscape (Ahrefs)
Organic SEO — Ahrefs Site Explorer
| Brand | Domain Rating | Org. traffic/mo | Org. keywords | Keywords Top 3 | Ref. domains | Backlinks | Org. value/mo |
|---|
Voxagent voxagent.ru | | | | | | | |
Zvonobot zvonobot.ru | | | | | | | |
Skorozvon skorozvon.ru | | | | | | | |
Twin24 twin24.ai | | | | | | | |
Voxys voxys.ru | | | | | | | |
👆 Click a brand row → per-brand SEO dashboard (authority, backlinks, top markets). Bars are normalized per column. Snapshot 2026-08-06.
AI Share of Voice
Voxagent holds 0% AI share of voice — zero impressions across ChatGPT, AI Overviews and AI Mode, while comparable competitors accumulate 47.9K impressions/mo.
Voxagent 0.0%Zvonobot 55.8%Skorozvon 19.8%Twin24 1.1%Voxys 23.2%
AI Visibility — Brand Radar (impressions / month)
👆 Click a brand row → popup with its AI citations (ChatGPT prompts whose answers cite that domain: prompt · volume/mo · co-cited domains). Brand Radar matches brand names, not domains, so common-word brands are flagged and excluded from the scale. Snapshot 2026-08-06.
GEO / AI visibility. Voxagent is cited by no AI engine and registers 0 impressions, while Zvonobot (26.7K impressions/mo) and Voxys (11.1K, plus 20 ChatGPT citations) already appear in AI answers for Russian voice-automation queries. Notably, Zvonobot earns impressions but no ChatGPT citations — it surfaces in AI Overviews without being quoted as a source, whereas Voxys and Skorozvon are quoted. That distinction matters: citation, not impression, is what puts a vendor into an AI-generated shortlist. With the category's Russian-language AI answer layer still thinly sourced, structured, citable content — pricing mechanics, integration guides, industry playbooks — is the fastest route in.
Funnel & GEO Benchmarks — B2B SaaS
Reference ranges for a B2B funnel — Organic visit → Lead → Sales-Qualified → Won — adapted from ESA's benchmark set (Ahrefs, Semrush, FirstPageSage, OpenView, Seer Interactive). Rows marked ROI-model input feed the forecast below.
§1 Conversion benchmarks
| Funnel step | Low | Healthy | Top | What moves it |
|---|
| Organic visit → Lead | 0.5–1% | 1.5–2.5% | 3–5% | Clear service pages, case studies, fast site, visible contact CTA |
| Lead → Sales-Qualified | 20–30% | 35–45% | 50–60% | ICP fit, fast follow-up, qualification, trust signals |
| SQL → Won | 10–18% | 20–30% | 30–40% | Proposal quality, expertise, security/compliance, references |
| Sales cycle | 1–3 mo | 3–6 mo | 6–12 mo | Longer for enterprise and regulated buyers |
| Avg. visits per page (peak) | — | 45 | — | ROI-model input. Ahrefs top-pages — trimmed mean of ranking pages. Median of Zvonobot's and Skorozvon's commercial pages — the closest comparable Russian-language competitors; discounted for a DR-0 domain with no ranking history |
| Customer value (LTV) | — | $8K | — | ROI-model input. Derived from the published Rost plan (₽150,000/mo) held ~12 months, converted at ~₽95/$1 and discounted for mix with the ₽50,000 Start plan. Replace with Voxagent's real ACV and retention. |
§2 GEO benchmark ranges
| GEO metric | Early | Healthy | AI-heavy | Notes |
|---|
| LLM traffic ratio | 0.01–0.04× | 0.05–0.12× | 0.15–0.30× | LLM referrals as a multiple of organic |
| LLM conversion boost | +50–120% | +300–500% | +400–700% | ChatGPT highest; AI Overviews lowest |
| Content success rate | 30–50% | 45–65% | 50–70% | Service pages and how-to content rank best |
| Months to peak | 3–6 | 6–14 | 14–24 | Per published cohort |
| Brand traffic multiplier | 0.05–0.15× | 0.20–0.40× | 0.45–0.90× | Fixed at 3.5× brand-lift in the model |
ROI Forecast — package × period
Cumulative Revenue vs Cost
Model = ESA ROI Calculator (cohort ramp → brand/LLM channels → funnel → cumulative ROI)
- Package sets pages/mo & monthly cost: Start 10/$5.5K · Growth 25/$10.95K · Scale 40/$15.5K
- Behavioural assumptions held at the Healthy tier of the benchmark tables above (months-to-peak 12, success 50%, LLM ratio 0.05, LLM conv ×2.0, visit→lead 1.0%, lead→won 8%, avg visits/page 45, brand amplification ×3.5)
- LTV $8K is an assumption and the dominant lever — Derived from the published Rost plan (₽150,000/mo) held ~12 months, converted at ~₽95/$1 and discounted for mix with the ₽50,000 Start plan. Replace with Voxagent's real ACV and retention.
- Brand amplification is fixed at ×3.5 per the ESA calculator reference. Combined with the assumed LTV it is what drives the headline figures: the model multiplies earned organic traffic by ~4.5× across brand and LLM channels before the funnel is applied. Treat the absolute revenue as an upper-bound scenario, and the shape (break-even month, package ranking) as the usable output.
- This domain starts from zero organic traffic, so every figure here is a modelled projection with no observed baseline to calibrate against — the widest uncertainty band of any input. Estimates on the stated inputs, not guarantees.