Paket Case Study
Paket Taxi worked with Webtures on a programme covering organic search and generative AI visibility together. For a courier and same-day delivery brand running bicycle, motorcycle and vehicle courier lines, the starting point was not content but access. Before a brand can appear in generative answers, the crawlers feeding those surfaces have to be able to reach the site at all.
Why Paket Taxi needed SEO and GEO support
Four structural challenges specific to the fast-delivery category shaped the brief.
- AI crawlers were shut out. The
robots.txtfile atpakettaxi.comdisallows generative crawlers, among them ClaudeBot, GPTBot, CCBot, Google-Extended, Bytespider, Amazonbot, Applebot-Extended and meta-externalagent, with a Content-Signal line readingsearch=yes, ai-train=no. The configuration most likely came from an infrastructure default rather than a deliberate decision; the consequence is the same either way, since without access no amount of content produces visibility on AI surfaces. - Brand queries dominated by job seekers. Searches containing the brand name are driven by employment intent rather than customer intent, with courier applications, earnings and working conditions accounting for much of the volume. That teaches generative engines to model the brand primarily as an employer rather than as a service provider.
- Local intent with no matching architecture. Courier demand forms at city and district level. The brand runs three offices, in Yenimahalle in Ankara, Muratpaşa in Antalya and Bornova in İzmir, but has no page structure describing service coverage by location, so that intent goes unanswered.
- A concrete gap in the trust layer. No privacy notice or data protection disclosure could be found on the site; the addresses tested returned 404. Google's quality rater guidelines treat trust as the most decisive part of E-E-A-T, and for a service model that collects personal data, the absence of these pages reads as a trust deficit rather than an oversight.
Goals set for the Paket Taxi project
The programme aimed to turn AI crawler access into a conscious decision, to establish the brand as the correct entity within the courier and rapid delivery category, to separate customer intent from employment intent, to give city-level service intent a page architecture capable of answering it, and to complete the missing trust signals. Because the category operates under Turkish postal services legislation and its authorisation regime, service descriptions were written to sit inside that frame.
How Webtures approached SEO and GEO for Paket Taxi
Webtures applied a strategy combining technical SEO, content architecture and generative visibility. The premise is simple: there is no separate mechanism for appearing in generative answers. Being crawlable, indexable and snippet-eligible is the precondition for being cited at all.
Technical SEO and AI accessibility
The first workstream was robots.txt. With generative crawlers disallowed, no volume of published content can make the brand citable on AI surfaces. Access policy was reviewed crawler by crawler, and the distinction between search crawlers and model-training crawlers was made explicit, since these are separate decisions. With access clarified, the work moved on to crawlability, indexation, page experience and structured data.
Content and semantic authority
Content was developed around the questions people actually ask. In this category they are transactional and time-sensitive: how to call a motorcycle courier, how same-day delivery works, what intra-city parcel sending involves. The same layer clarified the brand's category identity. Paket Taxi is a courier and rapid delivery brand, not a freight or logistics company, and that distinction matters to generative engines because a brand filed under the wrong category does not appear in answers to the right questions.
GEO and AI visibility
Work was carried out to strengthen presence across ChatGPT, Gemini, Perplexity and Google AI Overviews. A category-specific question set was defined and generative answers to it were monitored. The set was deliberately built along two axes, service intent and employment intent, so that questions where the brand appears as a service provider could be separated from those where it appears as an employer. Measurement relied on the Generative AI performance report in Search Console; third-party tools claiming access to Google's internal metrics were not accepted as evidence.
Experience and conversion
Page structures and journeys were arranged so visitors could reach the right action quickly. In this model the transaction completes on the phone or in the app rather than on the page, so the page's job is to shorten the handoff rather than to persuade. Two audiences competing for attention on the same screen was creating ambiguity, so the customer journey and the courier applicant journey were separated. The missing privacy and data protection pages were addressed in the same workstream.
What was carried out in the Paket Taxi project
Webtures delivered work across technical SEO, content architecture, site structure, structured data and GEO. The main workstreams:
- Detection of AI crawler access and a crawler-level review of
robots.txtpolicy - Crawl, indexation and page experience audit
- Entity clarification within the courier and rapid delivery category
- Separation of customer intent from employment intent
- City and service-area page architecture with structured business data
- Completion of trust pages, including privacy and data protection notices
- Question-set monitoring across generative answer surfaces
What the SEO and GEO work delivered
The programme was tracked across three reinforcing layers rather than a single number: organic search performance, local and service-area visibility, and citation in generative answers. The measurement frame combined Search Console, analytics and regular review of the defined question set.
Organic traffic and search visibility
Search visibility was reported by query intent rather than by volume alone. Service queries and employment queries were tracked separately, because read as one curve their sum makes the brand's commercial visibility look stronger than it is. On the local side, the measure was how much district-level demand actually mapped to a page capable of answering it.
Visibility across AI platforms
Generative answers to the defined question set were monitored, tracking separately where the brand was cited and where it was absent. This only becomes meaningful once crawler access is open; while access is closed, what gets measured is not the brand's content but its absence. Access policy was therefore reported as the precondition of the measurement.
The outcome of the Paket Taxi and Webtures partnership
The strategy addressed organic search performance and AI-assisted visibility together. In a category where the decision is made in seconds, visibility is built in a fixed order: technical access first, correct category identity second, content depth last. Break that order and the content does not lose its quality, it loses its reachability.