Real pushbacks, real handle scripts
Sourced from 29 prospect conversations. Each entry: what they said, why they said it, the handle script, proof points, and when to escalate.
Sourced from 29 prospect and customer conversations across BFSI, retail, FMCG, agency, streaming, gaming, agri, gifting and international SE Asia. The handle scripts are what closed the moment in the room, or, where we lost ground, what should have been said. Real weaknesses (pricing inconsistency, failed-generation billing, voice quality, PSD export, ORM accuracy at Bata) are flagged honestly with a non-defensive way to address them. Don't sell past them.
Category 1: Pricing & ROI Skepticism
"Your pricing is inconsistent, I've heard three different numbers from your team."
Heard from: Grihum, TAFE, EloElo, Winni, Mahindra Insurance Brokers · Severity: Critical
Why they're saying this: They've talked to two Zocket people, or compared notes with another prospect, and the numbers don't line up. This is a trust signal before it's a budget signal. They're wondering whether we know what our own product costs. Historically they were right to ask.
How to handle it: Own it, then give them the actual model:
"You're right to push on that, and I'm not going to improvise another number at you. The model is three parts: a platform subscription, the data collection, which scales with how much conversation your brand actually generates, and metered usage. Give me the platforms you want monitored and roughly your volumes, and I'll come back with a firm written number built from our cost model, not a guess."
Then book the follow-up and build the quote from the scraping calculator. Never quote a figure you're not certain is current.
Proof points to use:
- One model, three components: subscription + scraping + credits. Say the three words the same way every time, consistency itself is the answer to this objection.
- Scraping is computed from real per-platform unit costs, not a guess, which is why it differs by brand.
- Creative engagements are scoped in masters and adaptations, and iterations don't count against the cap.
When to escalate: Any written quote, multi-quarter commit pricing, white-label/OEM, or partner economics. Sales lead, always. See Pricing.
"What's the actual ROI? Show me before we commit."
Heard from: Matrimony, IIFL, Croma, EloElo, Mahindra Insurance Brokers, TAFE · Severity: Frequent
Why they're saying this: Procurement gates require a hard business case. They want time-saved, ROAS-uplift, or attribution-uplift numbers they can present internally, not feature talk.
How to handle it: Lead with one concrete reference: "Bharat Matrimony manages ₹2–2.5 crore monthly Google Ads spend through Performance AI. Their MIS team used to build the morning report manually with two people, now it's automated and they save 3–4 hours every day. Their root-cause diagnosis time on CPA creep dropped from 15–20 days to under one. Across our book we see customers cut reporting time roughly 80%." Then pivot to their world: "What does an hour of analyst time cost you, and how many CPA-creep events have you eaten in the last quarter?" Make them put a number on the status quo before you put one on the solution.
Proof points to use:
- Bharat Matrimony: $2–2.5M/month managed, 3–4 hours/day saved
- $120M/month managed across 250+ brand accounts
- IndiGo production case study
- Croma: 80% of conversions sit beyond Google's 7-day API window. They're losing visibility on the majority of their funnel today
- 80% reduction in time spent on data compilation/reporting (cross-customer benchmark)
When to escalate: When they ask for a written ROI guarantee or want to anchor commercials to a specific uplift KPI. That's a CFO-level conversation, not an AE one.
"How are AI/LLM credits actually counted? I don't want to be surprised by a bill."
Heard from: GQ Apparel, Matrimony, TAFE, EloElo · Severity: Frequent
Why they're saying this: They've been burned by usage-based SaaS. They want predictability and a meter, not an invoice surprise.
How to handle it: Do not walk them through per-query or per-generation arithmetic on a call. That is precisely what created the inconsistency problem above, and the numbers move. Instead:
"Usage is the metered part of a three-part model, subscription, data collection, and usage credits. We estimate it from your actual volumes rather than quoting a per-action rate, and consumption is visible in-product so it's not a monthly surprise. Give me your volumes and I'll forecast it in writing."
If they want a control rather than a number, there's a real answer: one customer has asked for the ability to cap consumption by switching off AI-generated insights, and that's the shape of the lever.
Proof points to use:
- Live credit balance in the top bar on every screen, usage is visible, not hidden
- Trial workspaces come with initial credits, so they can test before any commercial conversation
- Usage forecast built from their volumes, delivered in writing
When to escalate: Any request for a committed usage cap, a not-to-exceed clause, or per-unit contractual pricing.
"If a generation fails, you still charge me?"
Heard from: Grihum (explicit), implicit across pricing discussions · Severity: Critical
Why they're saying this: They heard "every generation counts including failures" from a rep and that's a non-starter. It's the single most loseable trust moment in the pricing conversation.
How to handle it: Acknowledge it directly, don't paper over: "I'll be honest, that line has been in our pricing sheet and we're actively fixing it. The intent in our roadmap is that failed jobs refund credits, full stop. Until that ships as a default, I'm going to commit to you in writing that we credit back failed generations on review, and we'll make sure your contract reflects that. Don't take 'every-gen-counts' as the final word." Then move on. Don't dwell.
Proof points to use:
credit-usageroadmap explicitly scopes a Failed-status filter that refunds credits- Until productized, this is handled contractually by sales
When to escalate: Always loop in sales lead before signing, get the refund clause into the SOW.
"How is this priced internationally? We're not in India."
Heard from: GQ Apparel (Thailand), YogaRenew (US) · Severity: Emerging
Why they're saying this: GQ was quoted $1,000/month platform license + consumption, first USD deal. They want to know if that's a real tier or improvised.
How to handle it: "For international we have a platform-license tier starting at $1,000/month plus AI consumption, same architecture, same connectors, plus 24/7 India-based support which is actually a positive for US and SE Asia timezones. For attribution-first markets like the US, we typically run a Performance AI POC on one channel before scaling. What's your current ad spend and which channels are you running?"
Proof points to use:
- GQ Apparel (Thailand) Performance AI POC on Meta, first international self-serve
- YogaRenew (NYC), $400–500K/month ad spend, 24/7 India support cited as differentiator
When to escalate: USD pricing isn't formalized on every sheet yet, confirm with sales lead before quoting.
Category 2: AI Trust & Quality
"We don't trust AI with our brand. How do you keep it on-rails?"
Heard from: Mahindra Insurance Brokers, Grihum, ICICI HFC, Prism Johnson, Amber, 5paisa · Severity: Critical
Why they're saying this: Brand managers have seen Jasper/Copy.ai/general LLMs produce off-brand output. They're not skeptical of AI in the abstract, they're skeptical that AI can model their brand without constant babysitting.
How to handle it: "Three things make this different. First, every brand gets a 1–2 day setup pass where we crawl your site, ingest your guidelines, and build a brand-specific knowledge graph, your palette, voice, mandatory disclaimers, negative keywords, and translation-vs-transliteration rules. Second, every output goes through an independent AI checker agent before delivery, scored on 8 dimensions, color, logo, typography, image style, copy, tech specs, sub-brand ID, layout, with a 0–100 score and a failure reason. Third, the Compliance Checker app audits every creative against your brand guide and rule packs, flags issues by severity, and your configured approval flows in AI Designer route each asset to the right sign-off before it ships. We can show you that flow live on a creative we generate from your own website in 15 minutes."
Proof points to use:
- Live demo: crawl prospect's site, generate 2–3 concepts in 15 minutes (closed Grihum)
- 8-dimension compliance scoring
- AI checker is an independent agent, not the same model that generated the output
- 100–130+ BFSI compliance prompts pre-built
When to escalate: If they want a written compliance audit or a sample of the checker output on rejected creatives, loop product.
"We tried Jasper / Copy.ai / Pencil / Pebblely, it didn't work."
Heard from: Pivotroots (implied, "10+ tools"), LuLu ("we already use AI tools"), YogaRenew, Amber · Severity: Frequent
Why they're saying this: They've been around the block. They've seen generic generators fail on brand specificity, on Indian-market context, on multi-language, or on volume. They're tired of evaluating shiny demos.
How to handle it: "Most of those tools are general-purpose generators, they don't know your brand and they don't measure their own output. We're a brand operating layer: same knowledge graph that generates the creative also moderates the resulting comments in ORM and measures the performance in Performance AI. So if a generated creative underperforms, that signal flows back into the next generation. It's a closed loop, not a one-shot. And for Indian-market work specifically, we do contextual (not literal) translation across all major Indian languages with ElevenLabs Indian voices, plus a regional character system. What didn't work on the last tool you tried, was it brand fidelity, language, or volume?"
Proof points to use:
- Closed feedback loop: creative → distribution → performance → next generation
- 250+ brand accounts, $120M/month managed
- Brand-specific knowledge graph (not generic prompts)
- Indian voices via ElevenLabs, contextual translation per term
When to escalate: If they cite a specific competitor we've lost to before, flag in the CRM and brief sales lead before next call.
"The voice/video quality isn't good enough, feels flat / wrong season / wrong geography."
Heard from: Pivotroots (voice modulation flat, winter clothes in May campaign), YogaRenew (NRI vs Indian content) · Severity: Common, real weakness
Why they're saying this: They tested an output, and a specific thing failed. Voice modulation reads flat. A May campaign showed winter clothing. NRI content looked off for Indian buyers.
How to handle it: Don't defend, concede and route: "You're calling out two real gaps. On voice, the default modulation has been flagged before and we've started shipping custom-voice uploads via ElevenLabs so you can use your own brand voice instead of the default. On seasonal accuracy, that's a current weakness, the AI checker agent is being extended with a seasonal-mismatch gate, and the festival-kit feature in the knowledge graph is where we'll enforce that going forward. For now, my honest recommendation is to keep a quick human seasonal QA in your workflow until that ships, most agencies do, and we don't pretend otherwise."
Proof points to use:
- ElevenLabs custom-voice upload as a workaround
- Festival-kit / seasonal-gate on the AI checker roadmap
- Acknowledge the gap honestly, Pivotroots respected that more than a defense
When to escalate: Agency buyers especially, if voice or seasonality is a deal-breaker on a near-term campaign, flag for product. This is a known gap to product leadership.
"Are you sure the data is accurate? AI hallucinates."
Heard from: EloElo (explicit anti-hallucination concern), 5paisa, IIFL · Severity: Common
Why they're saying this: They've read about LLM hallucination and they don't want a dashboard that confidently lies about ROAS.
How to handle it: "We've designed for this from the start. There's a custom orchestration layer over the LLMs, Claude/Opus handles coding and dashboard tasks, Gemini handles reasoning, fine-tuned open models for specific jobs, with a brand-specific knowledge graph that grounds every answer in your actual data. The system also corrects unrealistic prompts: if you ask for a 10× ROAS, it'll surface what's realistic given your historicals and propose alternatives rather than making something up. We're running this across 20+ major Indian brands with no data hygiene issues, including $2–2.5 crore/month spend brands like Bharat Matrimony. Want me to demo a prompt that would normally trip a generic LLM and show you the guardrail?"
Proof points to use:
- Multi-LLM routing (Claude/Opus + Gemini + fine-tuned open) via Deep Agent
- Brand-specific knowledge graph as grounding layer
- Smart prompt correction (catches unrealistic 10× ROAS asks)
- 20+ major Indian brands in production, $120M/month spend
When to escalate: If the CTO is in the room and wants a system-architecture deep-dive, get Karthik or the engineering lead on a follow-up.
Category 3: vs. Competitors & Existing Stack
"How is this different from the other AI creative tools we've evaluated?"
Heard from: LuLu, Mahindra Insurance Brokers, Pivotroots, Prism Johnson, YogaRenew · Severity: Frequent
Why they're saying this: This is the table-stakes differentiation question. If you answer with feature-list, you lose. They want the one-sentence reason we win.
How to handle it: "Three things separate us. One, brand-specific knowledge graph: we model your brand before we generate, so output stays on-brand without prompt engineering. Two, multi-LLM orchestration with an independent AI checker that scores every output against your brand and compliance rules. Three, we're end-to-end: the same platform that generates the creative also measures it in Performance AI and moderates the comments in ORM AI. Point tools generate; we operate. Which of those three matters most to your team?"
Proof points to use:
- Brand knowledge graph (vs generic prompt-based tools)
- Independent AI checker agent (separation of duties)
- End-to-end loop: Creative AI → Performance AI → ORM AI on one knowledge graph
- 250+ brand accounts, $120M/month managed
When to escalate: If they name a specific competitor (Triple Whale, Genspark, Canva Enterprise, Jasper), flag the named competitor in CRM and prep talk-track for next call.
"We're evaluating Triple Whale for attribution. Why you over them?"
Heard from: YogaRenew (explicit, evaluating Triple Whale same day) · Severity: Common, for attribution-led buyers
Why they're saying this: Triple Whale is a real, focused attribution product. For complex multi-week, multi-touch journeys it's earned the reputation it has.
How to handle it: Be honest about coexistence: "For pure multi-week, multi-touch attribution Triple Whale is a solid product and we don't compete head-to-head with them. We coexist. Our wedge is different: we're the operating layer that takes the attribution signal and acts on it, creative regeneration when fatigue hits, comment moderation on the resulting ads, brand-compliant variant generation, cross-platform optimization recommendations. If attribution is your #1 unsolved problem and nothing else, Triple Whale may be the faster answer. If you also need creative pipeline, brand governance, and an AI command center across paid media. That's our shape, and we can plug Triple Whale's data in alongside us." Then ask what their #2 and #3 problems are, if attribution is the only one, lose gracefully.
Proof points to use:
- We coexist with attribution tools (Triple Whale, AppsFlyer, Branch, Singular all supported as connectors)
- Honest acknowledgment: Triple Whale is good at what it does
- Wedge: closed creative → distribution → ORM loop on one knowledge graph
When to escalate: If attribution is the only pain, let it walk. Don't try to outpitch a focused attribution tool on its home turf.
"Will this overlap with our Microsoft Copilot / enterprise AI investment?"
Heard from: 5paisa (explicit) · Severity: Emerging, will recur for any Dynamics/Office shop
Why they're saying this: Their CIO already wrote a check to Microsoft. They don't want to explain a second AI platform to procurement.
How to handle it: "Microsoft Copilot is a horizontal productivity AI, emails, documents, Office. We're a vertical marketing-and-brand-operations platform. We integrate with Dynamics 365 as a CRM connector, we're not trying to replace document or email workflows Copilot handles. The two complement each other: Copilot makes your knowledge workers faster on internal docs; Zocket makes your marketing team faster on creative, performance, and customer response. If procurement asks for a side-by-side, position us as enabling Copilot's CRM data through marketing intelligence, not displacing it."
Proof points to use:
- Microsoft Dynamics 365 native CRM connector
- Horizontal vs vertical framing
- BYOK option (use your own Microsoft/Anthropic/Google keys for compute)
When to escalate: If their CIO is gating, get a joint conversation with sales lead. This is a procurement narrative as much as a product one.
"Our agency already does this for us."
Heard from: Amber, Pivotroots, LuLu, Shalimar, Prism Johnson · Severity: Frequent
Why they're saying this: They've got a multi-year agency relationship and they're not looking to fire anybody. Pitching replacement gets you walked.
How to handle it: "We don't replace the agency, we replace the manual layer your agency does. Amber told us directly: 'the resize-and-adapt work is the manual part of my business.' That's where we land. Your agency keeps the brief, the strategy, and the creative judgment. We remove resize, language adaptation, format variants, compliance checks, fatigue detection, and reporting, the 80% of grunt work around the strategic 20%. Some agencies actually run Zocket inside their delivery, Pivotroots, simplemagic, gigatattva, AquaOrange in Thailand. Would it help to position this to your agency as their efficiency play, not a threat?"
Proof points to use:
- Amber quote: "the resize-and-adapt work is the manual part of my business"
- Tensai, simplemagic, gigatattva, AquaOrange are active agency-channel partners (not enemies)
- Agencies use Zocket as a force multiplier, not a replacement
When to escalate: If their agency is in the room or has veto, propose a joint demo with the agency, make them an ally, not an obstacle.
"We already have BigQuery / Tableau / our own data warehouse. Do we even need you?"
Heard from: EloElo (BigQuery), Pivotroots, GQ Apparel (own warehouse), Matrimony (Tableau) · Severity: Common
Why they're saying this: They've already invested in data infra. They're worried about being asked to re-pipe data into our system or building another silo.
How to handle it: "You don't re-pipe anything. Our BigQuery connector is live, you select your database and table the same way you'd select an ad account. We layer Performance AI's analysis on top of your warehouse, not in place of it. Tableau lives downstream, we can push views back, or you keep Tableau for executive reporting and use our prompt-driven dashboards for daily ops. The value isn't another data layer; it's the AI agent that talks to your existing data in natural language and tells you what to do."
Proof points to use:
- Google BigQuery connector live (EloElo, Pivotroots)
- 90+ connectors total, including bidirectional Google Sheets
- Matrimony plays well with their Tableau setup
- Process-only mode available, we can analyze without storing
When to escalate: If their data team wants an architecture review, route to engineering. Don't try to hand-wave.
Category 4: Implementation, Integration & Timing
"How long will this take to implement? We don't have engineering bandwidth."
Heard from: Grihum, ICICI HFC, IIFL, EloElo, Matrimony · Severity: Frequent
Why they're saying this: They've been burned by 6-month enterprise rollouts. They want to know there isn't a hidden engineering tax.
How to handle it: "Brand setup, knowledge graph build, brand rules, guidelines ingestion, is 1–2 days. Connectors are plug-and-play via OAuth, no engineering on your side. The standard motion is: NDA → paid POC on real data → one-week on-site assisted onboarding, Jetin and Rakesh did exactly this at IIFL, → gradual rollout starting with one ad account or sub-brand. Custom dashboards usually land within 24 hours of request during onboarding. The Matrimony playbook is the textbook: start with one Google Ads sub-account, prove value, then scale. Where would you want to start, one channel, one sub-brand, or one campaign type?"
Proof points to use:
- IIFL: NDA done → on-site one-week onboarding (Jetin + Rakesh)
- Matrimony: started with one Google Ads sub-account, scaled from there
- 1–2 day brand setup
- Custom dashboards in 24 hours during onboarding
When to escalate: If they want a guaranteed onboarding SLA in the contract, route to sales lead.
"Will integrating with our CRM be a nightmare? We're mid-migration."
Heard from: IIFL (Zoho → Dynamics 365), 5paisa (Dynamics), Croma, Matrimony, Pivotroots · Severity: Common
Why they're saying this: Many enterprise buyers are between CRMs. They want to know we won't block on a moving target.
How to handle it: "We support Zoho, HubSpot, Salesforce, Microsoft Dynamics 365, and Agile CRM natively. If you're mid-migration, which IIFL was, going Zoho → Dynamics. We can run manual upload as the interim path while the new CRM stabilizes, then switch to live connector when ready. You don't lose months waiting for your CRM stack to settle. What's your target CRM and timeline?"
Proof points to use:
- Native connectors: Zoho, HubSpot, Salesforce, Dynamics 365, Agile CRM
- IIFL Zoho → Dynamics 365 mid-migration, interim manual upload
- 90+ total connectors
When to escalate: Custom CRM (proprietary or non-listed), engineering scoping conversation.
"What about Blinkit / Zepto / Amazon Ads, we live on marketplaces."
Heard from: Pivotroots (Blinkit/Zepto, Amazon Seller Central) · Severity: Common for D2C, real weakness
Why they're saying this: A D2C brand on quick-commerce or Amazon has 40–70% of revenue flowing through those channels. If we can't read those, we're blind on most of their P&L.
How to handle it: Be honest: "Quick commerce, Blinkit, Zepto, currently needs manual workarounds, no native API yet. Amazon Ads and Seller Central are similar, marketplace complexity is a known gap on our roadmap. If your primary channel is quick-commerce or marketplace, I'm not going to overpromise, we'd start with a manual-upload pilot for that data and run the full Performance AI loop on your Meta and Google spend where we're strongest. Once native marketplace connectors ship, we'd add those in. Honest fit conversation: what % of your spend is on marketplaces vs Meta/Google?"
Proof points to use:
- Direct Excel/manual upload as interim integration
- Native connectors for Meta, Google, Snapchat, TikTok, DV360, AppsFlyer, Branch, Singular
- Honest acknowledgment beats a future-promise, Pivotroots respected it
When to escalate: If marketplaces are >50% of their spend, this is a product gap to flag to leadership, and likely not a fit today.
"We need cross-platform frequency capping across Meta + Google + programmatic."
Heard from: IIFL (explicit blocker for onboarding kickoff) · Severity: Critical for IIFL
Why they're saying this: A real BFSI buyer has a real waste problem, same user hit 12 times across channels. They want enforcement, not just diagnosis.
How to handle it: "Honest answer: we can diagnose cross-platform frequency overlap today, Meta + Google + programmatic, and surface it as an insight in the AI command center. Enforcing it as an automatic cap across platforms is not productized today. The current workflow is: we surface the overlap, recommend a budget shift, and a human approves the action on Meta and Google. If full programmatic enforcement is a hard requirement. That's a custom-scope conversation we'd want to scope with engineering, flag it now rather than promise it out-of-the-box."
Proof points to use:
- Diagnosis-side: live today, multi-channel overlap surfaced
- Enforcement: roadmap item, custom-scope if blocker
- Don't oversell, IIFL flagged this explicitly as a requirement
When to escalate: This is a real product gap. Flag to product leadership when surfaced; route enterprise asks to engineering scoping.
"Can we start small and expand? We're not committing the whole org day one."
Heard from: Matrimony, IIFL, Croma, Shalimar (3–6 month pilot) · Severity: Frequent
Why they're saying this: They want to de-risk. Internal politics also matter, they need a win to justify expansion.
How to handle it: "That's our playbook. Bharat Matrimony started with one Google Ads sub-account before scaling across the MCC. IIFL is starting with one ad account before adding DV360 and programmatic. The structure is: paid POC on real data with defined KPIs → one ad account or sub-brand goes live → prove the win → expand. We actually prefer this, it lets your team validate our reasoning before betting more spend, and lets us tune the brand knowledge graph against real data before scaling. Define a 90-day success metric with me right now and we work backward from it."
Proof points to use:
- Matrimony gradual-rollout playbook
- IIFL one-ad-account-first kickoff
- Paid POC is the standard motion, not free trial
- Custom dashboards in 24 hours during onboarding
When to escalate: Always, KPI-bound POC contracts should be reviewed by sales lead.
Category 5: Brand Safety, Compliance & Data Governance
"We're in a regulated industry, can you handle compliance approvals on every creative?"
Heard from: 5paisa, IIFL, ICICI HFC, Grihum, Mahindra Insurance Brokers · Severity: Critical for BFSI
Why they're saying this: RBI, SEBI, AMFI, exchange-approval workflows are hard regulatory gates. One non-compliant creative going live is a fine and a headline.
How to handle it: "This is our strongest BFSI play. The Compliance Checker and approval flows enforce required disclaimers, RBI fair-practice, SEBI ad code, AMFI standard warnings, plus prohibited phrases, restricted testimonials, and the Google Ads financial-disclosures + Meta special-ad-category routing. We have 100–130 pre-built BFSI compliance prompts. Every output is scored before delivery; anything below threshold goes to the compliance reviewer queue with a provenance panel showing which rule it failed. For 5paisa specifically, the exchange-approval workflow is the textbook case. We shorten the loop but don't replace the exchange reviewer. Chola Investments is our BFSI reference if you want to see this in production."
Proof points to use:
- 100–130 BFSI compliance prompts pre-built
- 8-dimension compliance scoring with provenance
- Chola Investments reference (BFSI)
- IIFL onboarding kickoff in progress with these guardrails
When to escalate: For exchange-approval workflows or RBI/SEBI-specific edge cases, get product + compliance specialist on the call.
"Where does our data sit? Can you process without storing?"
Heard from: IIFL, 5paisa, EloElo · Severity: Critical for BFSI
Why they're saying this: Data residency is a hard gate. They want to know you can pass a CISO review.
How to handle it: "Yes, we support a process-only mode for privacy-sensitive clients where we don't persist the underlying data, only the analysis output. Enterprise deployments can use your own API keys for OpenAI, Anthropic, and Google, BYOK, so you control your model access logs and billing. For the highest-trust clients we deploy on your own cloud infrastructure. Default deployment is our cloud with role-based access and encryption at rest. What's your CISO's red line, process-only, BYOK, or own-cloud?"
Proof points to use:
- Process-only mode
- BYOK for OpenAI/Anthropic/Google
- Client-cloud deployment option for highest-trust
- IIFL/5paisa BFSI deployments in this category
When to escalate: Own-cloud deployment requires engineering scoping. Loop sales lead.
"Do we own the creative outputs? Any IP issues?"
Heard from: ICICI HFC (explicit) · Severity: Common (assume universal)
Why they're saying this: They've read about training data and copyright suits on generative AI. They want a clean IP answer.
How to handle it: "Yes, all outputs are exclusive to your brand. No shared training across customer assets, no syndication of your creative to other clients, no IP conflicts. Brand-specific knowledge graphs are siloed per tenant. We can put that in the contract."
Proof points to use:
- Per-tenant siloed knowledge graphs
- Exclusive IP on outputs, contractually
- No cross-customer training
When to escalate: Custom IP/IPR language in the MSA, route to legal.
Category 6: Team, Change Management & Engagement Model
"Will this replace our creative team / our marketing team?"
Heard from: Amber, Pivotroots, LuLu, Shalimar · Severity: Common
Why they're saying this: Team morale and internal champion management. If reps walk in saying "fire your designers," they walk out the door.
How to handle it: "No, and we don't pitch it that way. Zocket replaces the manual execution layer, resize, language adaptation, format variants, compliance checks, fatigue detection, reporting, not the strategic thinking. Your team still owns the brief, the strategy, and the final creative judgment. We just remove the 80% of grunt work around the strategic 20%. The teams we work with typically reallocate time from production to higher-order work, Bharat Matrimony's MIS team got 3–4 hours/day back, which they now use on diagnostic deep-dives instead of building the same report from scratch every morning."
Proof points to use:
- Bharat Matrimony: 3–4 hours/day reallocated, not eliminated
- Amber quote on adaptation as the manual layer
- We don't replace strategy/judgment, we remove execution drag
When to escalate: If a head of design or head of marketing is openly hostile, get a joint session with sales lead to reframe the value to their team specifically.
"Do we want managed-service or self-serve? Help me pick."
Heard from: Mahindra Insurance Brokers (platform), GQ Apparel (self-serve explicit), ICICI HFC, TAFE, EloElo · Severity: Frequent
Why they're saying this: They don't fully understand the two motions and they want a guided answer, not a sales pitch.
How to handle it: "Two motions, same pricing architecture underneath, subscription, data, usage. Managed-service adds a Zocket team who handle briefs, build dashboards, run ongoing optimisation and do creative QA, so you get delivered outputs. Platform-only / self-serve gives your team the full UI and they operate it. The split we're seeing: BFSI and enterprise buyers, Mahindra, IIFL, ICICI, lean platform-only because they have internal teams. Agencies and high-volume creative shops, Pivotroots, Winni, lean managed because they want the output delivered. International self-serve is growing, GQ Apparel in Thailand was our first explicit PLG POC. Which way is your team set up?"
Proof points to use:
- Mahindra, IIFL, ICICI HFC → platform-only
- Pivotroots, Winni → managed
- GQ Apparel → self-serve international
- The difference is a services layer, not a different pricing model, say that plainly
When to escalate: Hybrid asks (managed creative + self-serve performance), sales lead to scope.
"Do we get a dedicated team / what timezone is support?"
Heard from: YogaRenew (US), ICICI HFC, Bata · Severity: Common
Why they're saying this: Enterprise buyers want a name and a number. US clients specifically want timezone coverage.
How to handle it: "Enterprise customers get a dedicated implementation resource for setup plus a daily or weekly cadence call during the first 30–60 days. Bata is on a weekly catchup; IIFL got on-site for one week. Standard support is 24/7 from India, which is actually a positive for US clients on east-coast hours, your morning brief is built while you sleep. Slack alerts, email reports, and shareable dashboard links are the async default."
Proof points to use:
- IIFL: one-week on-site (Jetin + Rakesh)
- Bata: weekly catchup cadence (after daily in early stage)
- 24/7 India support, YogaRenew differentiator
- Slack + email + shareable links
When to escalate: If they want a written support SLA with response time guarantees, sales lead.
Category 7: Industry / Use-Case Fit
"Your demo is BFSI / D2C / retail, does it work for our industry?"
Heard from: Shalimar (FMCG food), Prism Johnson (building materials), TAFE (agri machinery), GQ Apparel (apparel/Thai), LuLu (multi-region retail), YogaRenew (US education) · Severity: Frequent
Why they're saying this: They want to see something that looks like their world before they commit.
How to handle it: "Fair, let me find the closest reference. We're in production across BFSI (Chola, IIFL onboarding, ICICI HFC pilot), streaming (Zee), gaming (EloElo), matrimony (Bharat Matrimony at ₹2–2.5 crore/month), retail (Croma, Bata, Lenovo sub-brands), gifting (Winni), agency (Pivotroots, Amber), apparel international (GQ Apparel Thailand). Even if your exact vertical isn't here, the mechanics are the same, brand knowledge graph plus AI checker plus closed loop. The fastest way to prove it is to crawl your website live in this call and generate 2–3 creative concepts in 15 minutes. Want to do that now?"
Proof points to use:
- 250+ brand accounts across verticals
- BFSI cluster (Chola, IIFL, ICICI HFC, Grihum, 5paisa, Mahindra)
- Multi-vertical: Zee, EloElo, Matrimony, Croma, Bata, Winni, Pivotroots
- Live website-to-creative demo in 15 min (closed Grihum)
When to escalate: If they want a written portfolio under NDA, route to sales. We have one for BFSI specifically.
"Can you do consumer research like Kantar / Ipsos?"
Heard from: Amber, LuLu · Severity: Emerging
Why they're saying this: They're testing whether we overpromise. Some buyers will conflate AI consumer-insight with primary research.
How to handle it: "No, and we say so explicitly. Zocket uses secondary data, online sources, social signals, search, ad libraries, not primary research panels like Kantar. We do consumer-insight tracking, sentiment analysis, audience demographics, and content-gap analysis through AI Consumer Research and the AI Researcher. If you need primary research with recruited panels. That's Kantar or Ipsos and you should bring them in alongside us. We complement, we don't replace."
Proof points to use:
/consumer-insights/social, secondary-data sentiment, share of voice + demographics- Honest scope: no panels, no primary research
- We coexist with Kantar/Ipsos
When to escalate: None, this is a clean honest scope conversation. Don't oversell.
"Can your AI handle a non-marketing use case, internal comms, training, events?"
Heard from: ICICI HFC (200+ statics, mostly training/internal/event), Grihum (HR, policy, branded merch) · Severity: Emerging
Why they're saying this: They have heavy non-marketing creative volume. The competitive set shifts (Canva Enterprise territory).
How to handle it: "Yes, Creative AI generates static and video creative for any use case, not just paid media. ICICI HFC's 200+ statics/month includes training collateral and internal comms. The brand knowledge graph enforces consistency the same way it does for marketing, same compliance checking, same approval flows, same multi-language support. Where this matters most is brand-consistency across departments: your training content and your campaign content read like the same brand. Want to scope a non-marketing pilot too?"
Proof points to use:
- ICICI HFC: 200+ statics/month, mostly training + internal + events
- Grihum: HR, policy, branded merchandise, regional greetings
- Compliance checking + approval flows apply to all creative, not just paid media
When to escalate: If volume is >300/month internal, pack pricing conversation with sales lead.
Category 7b: New Patterns from Recent Calls
"We need category intelligence, what creatives usually work for our industry. Not just AI generation."
Heard from: Piramal Consumer Healthcare (explicit) · Severity: Emerging, likely to recur for CPG / FMCG buyers
Why they're saying this: They're not hiring a production vendor. They're looking for a strategic partner who understands their category norms (what hooks work for face wash, what formats win for sunscreen CPA campaigns) and can translate a brief into evidence-based creative routes, not just execute whatever prompt they give.
How to handle it: "That's a great challenge for us because it's exactly how we're designed. Two inputs drive what we recommend: first, competitor ad audits from Meta Ad Library and public creative trackers, we'll audit which formats and hooks your top 3–5 competitors are running and how they're performing. Second, your own historical performance data once you connect your ad accounts. We identify which hooks, formats, and layouts have actually worked for you, and we use that to brief the creative generation. So when you brief us on a Lactocalamine CPA campaign for Meta + Blinkit. We come back not with one option but with multiple creative routes ranked by category evidence, emotional, functional, UGC-style, testimonial, so you choose based on your brand POD, not guesswork."
Proof points to use:
- The Competitive Cloner and AI Researcher audit competitor ads (format, hook, performance signal)
- Performance AI connects ad accounts to identify winning hooks and formats from your own data
- Competitive + owned-data inputs create brief-to-routes, not brief-to-one-output
- Piramal: generated Lactocalamine creatives from product listing page alone in under 5 minutes
When to escalate: If they want a written category report before committing, route to content/product team. Don't improvise category benchmarks on the call.
"We're a B2B brand, does AI creative even work for us?"
Heard from: Signode India (explicit initial skepticism) · Severity: Emerging, any industrial / B2B buyer
Why they're saying this: They associate AI creative with lifestyle UGC and consumer ads. B2B has machinery, technical specs, and conservative buyer audiences. They're not sure AI can produce exhibition graphics, distributor comms, or product-capability videos without looking cheap.
How to handle it: "B2B creative has actually been one of our stronger early use cases. The knowledge graph models your products the same way it models a consumer brand, specifications, key use cases, target audience (distributors vs end buyers vs installers), brand tone, and any compliance or certification messaging. We generated Earth Day campaign concepts and in-store/office branding for Numeric UPS that looked product-accurate and on-brand without any stock imagery. For B2B the value is often in volume: 8–10 creatives/month at agency TAT and cost when you're doing product launches, exhibition stand design, and distributor adaptation in parallel. Want to share a product link and we'll generate one sample right now?"
Proof points to use:
- Numeric UPS (B2B electronics): Earth Day video, in-store branding, wall art generated via Zocket
- Product-accurate generation from URL / spec sheet, not just lifestyle prompts
- Multilingual support: Arabic, Thai proven; Japanese/Korean/Chinese/Malay in roadmap
- Same TAT applies: same-day statics, 24-48 hours video
When to escalate: If they have complex technical imagery needs (CAD renders, animated explainer of mechanical processes), engineering/creative team scoping conversation.
"I want to see this work for fashion / premium imagery, AI tends to look off."
Heard from: ABFRL (explicit concern about quality and printability at large format) · Severity: Emerging, fashion, luxury, and premium retail
Why they're saying this: Fashion brands have trained eye for lighting, fabric texture, model quality, and proportionality. Early AI creative tools produced generic-looking outputs that weren't print-ready or high-fashion enough. Print at large display sizes (store signage, OOH) is a specific technical requirement, pixelation is a real deal-breaker.
How to handle it: "Fashion is a genuinely higher bar and we don't pretend otherwise. A few things matter here: first, we use high-resolution generation models with upscaling specifically for large-format print, the output is dimensioned for your requested print size, not scaled up from a web image. Second, you bring your own product images as the reference, the AI doesn't invent the fabric or the model, it dresses your product in the right context. For ABFRL specifically we generated carousel and UGC-style samples using their public product images in the demo, and the output retained the product detail. Third, copyright, everything generated is commercially licensed, IP transfers to you, and we can put that in the contract. The honest bar: if you need TVC-grade fashion cinematography today, we're a complement to your agency, not a replacement. But for social, digital, regional adaptation, format resizing, and seasonal variants, same-day delivery that an agency charges 2-week TAT for, we're strong."
Proof points to use:
- High-res generation + upscaling for large-format print
- Product-image-anchored generation (not AI-invented garments)
- ABFRL carousel and UGC-style sample in demo using their own product images
- Commercial licensing + IP transfer contractually
- Best fit: social, digital, regional adaptation, format variants; not TVC-grade cinematography
When to escalate: If they need print-readiness guarantees for OOH at billboard scale (4m+), engineering/creative team should review the upscaling specs before commitment.
Category 8: Product Gaps to Watch (Sales Leadership Flag)
These are objections where we don't yet have a clean answer. Flag for product/sales leadership when they surface. Don't paper over.
"Where's PSD / AI / CDR / native print export?"
Heard from: Amber (PSD/AI), Pivotroots (PSD), Grihum (CDR), Mahindra Insurance Brokers (print PDF) · Severity: Critical for agencies + print-heavy buyers
Honest handling: PNG, JPEG, and PDF (print) are standard today. Native PSD upload and AI/CDR export are not GA. "PSD Upload Support" is on the near-term roadmap. For now, position upscaling-from-compressed-source as the print quality story; route open-format requests through implementation, not the demo. Don't promise PSD on the call.
Flag: This is now a recurring deal-blocker for agency and BFSI/NBFC buyers. Product leadership needs a ship date.
"ORM gets responses wrong / privacy masking breaks the workflow."
Heard from: Bata (production issues) · Severity: Critical, live customer
Honest handling: Acknowledged production gap. "Thank you" comments wrongly triggering ticket creation, privacy masking making responses ineffective, ticket creation failing intermittently. These are actively being fixed in current sprints. Don't sell the auto-response quality bar without acknowledging human review is needed on positive/appreciation cases. Bata is on weekly cadence to monitor.
Flag: ORM AI accuracy at production scale is a customer-success risk. Don't pitch full auto-response to new ORM prospects until fixes ship.
"Auto-apply changes to Meta/Google, can you push, not just recommend?"
Heard from: Saurabh (system-user token / ads_management scope), implied across Performance AI calls · Severity: Emerging, power-user wedge
Honest handling: Today we recommend; a human approval pushes the action. Full programmatic execution via Meta/Google API is roadmap, not productized. Temporary data-fetch tokens work; ads_management execution scope is not enabled on Zocket-managed accounts today. Saurabh wants to scale ₹30–40L → ₹3 crore/month and is gated on this, high-value bug fix.
Flag: API/developer-tier offering needs a product decision, service or feature. Loop CTO + product.
"Organic engagement uplift, not paid media, what's your story?"
Heard from: LuLu Group India · Severity: Emerging
Honest handling: We don't have a strong productized story for organic-engagement uplift yet. Our pitch is built around paid-media outcomes (ROAS, CTR, reporting), brand governance, and ORM. If a buyer's #1 ask is organic-social engagement. That's a soft fit, be honest, propose a content-engine pilot rather than overpromising an "engagement uplift" outcome.
Flag: Organic-engagement narrative gap, product leadership decision: build or de-prioritize this segment.
"Prompt-driven journey orchestration (CDP), does it actually do this end-to-end?"
Heard from: 5paisa (the "recover ₹10cr from dormant customers" prompt example) · Severity: Emerging
Honest handling: The prompt-to-cohort-to-journey-to-content claim made to 5paisa is at the aspirational edge of what's productized in [[retention-ai]] today. Lightweight CDP + CleverTap/AppsFlyer integration is real; fully automated cohort + journey + content from a single prompt is partly aspirational. Don't replay that specific pitch verbatim until product confirms the end-to-end flow.
Flag: Confirm with product where retention-ai is GA vs aspirational before in-person Mumbai 5paisa follow-up.
That's the canonical objection set. When a prospect raises a pushback not on this list, log it back to the wiki, recurring objections are the signal we use to expand this guide. When a category 8 gap blocks a deal, escalate to product/sales leadership the same day; don't let the rep absorb the loss alone.