Reading between the lines: AI and customer acquisition in professional services
Bibliographic record
Abstract
Purpose Professional service firms face complex challenges in acquiring new clients, owing to the high degree of trust, customization and client involvement these services require. This study aims to explore how large language models (LLMs), such as ChatGPT, can analyze early-stage text-based communications to identify linguistic markers that predict purchase readiness. By focusing on the professional services domain, the study contributes to service marketing theory by introducing a novel method to assess service interest through client-generated text. Design/methodology/approach For two years, the authors collected responses submitted through an inquiry form for a home improvement firm. The artificial intelligence (AI) platform ChatGPT was trained to assess the level of specificity in the language used by prospective customers and generated a specificity score for each response. These scores were then used to predict the likelihood of customer purchase over time using an accelerated hazard model. Findings The results indicate that the specificity scores generated by ChatGPT effectively predict a customer’s position within the sales funnel and likelihood of purchase over time. Customers who provided more detailed information, as measured by AI, exhibited a higher probability of conversion. Practical implications This study provides actionable insights for managers aiming to optimize customer acquisition efforts and minimize resource waste. It demonstrates how AI – specifically LLMs – can be leveraged to analyze unstructured text from prospective customers and identify linguistic signals (e.g. specificity) that are predictive of purchase likelihood. Originality/value Consumers reveal valuable insights through the language they use. While it has traditionally been difficult to empirically analyze this verbiage, the emergence of LLMs enables the transformation of qualitative text into measurable indicators. These tools allow firms to make data-driven predictions about customer behavior. This study introduces a novel methodological approach to the analysis of customer acquisition.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.002 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".