How's your internet?

A guest blog post by Dustin Gilbreath and Hans Gutbrod

If you are reading this, maybe your internet isn’t so bad. Maybe it took some time to load this page and while waiting for the page to load, you thought, “It’s bad.” The quality of internet service is a frequent topic of conversation in Tbilisi, where access to the Internet is the highest in Georgia, and satisfaction with the service varies widely. This has motivated a quick online poll – undertaken by Dr. Hans Gutbrod, in March 2015 – on assessments of internet service in Tbilisi. While not a proper random survey, such a poll (conducted with Google Forms) can still be informative, though it will primarily provide the views of people that have time and feel strongly enough to fill out the survey. The questionnaire was distributed on the Megobrebs listserv, newsgroup and via twitter, and was administered in English only. A total of 52 people based in Tbilisi filled out the survey.

What then are the findings?

Thirty four respondents reported experiencing problems with their home internet access, with 16 complaining about regular short interruptions in service; 9 about crawling internet, 10 about outages lasting longer than 5 minutes, and 13 having trouble connecting with multiple devices.

Overall, respondents’ satisfaction with internet service, measured by their willingness to recommend the service to a neighbor, was medium, with a mean of 2.92 points on a five-point scale, with code ‘5’ meaning not willing and code ‘1’ meaning willing to recommend. With all the caveats mentioned above, the numbers seem to suggest that Caucasus Online customers appear to be more satisfied than Silknet subscribers. The latter were slightly over half a point (0.63) less willing to recommend their internet service provider (ISP) to a neighbor willing to pay the same service fee, indicating a lower level of satisfaction. Caucasus Online subscribers also rated the customer service they had received in instances where they encountered a problem 0.74 point higher on average compared with Silknet subscribers.

Note: Respondents’ answers were recorded using five-point scales. For the first question on the chart above, code ‘5’ corresponded to very bad service and code ‘1’ – to very good service. For the second question, code ‘5’ corresponded to the answer ’not willing to recommend the [respective] service to a neighbor’ and code ‘1’ – to the answer ‘willing to recommend the service to the neighbor’.

Interestingly, satisfaction with customer service varies by language used by the respondent to communicate with the company. The 9 respondents who only spoke with their ISP in English rate the service almost a point higher than those who communicated with the ISP in Georgian (19 respondents). The 18 respondents reporting using a combination of languages (Georgian, Russian or English) fall in between the two. Hence, Georgian speakers appear to be the least satisfied with their ISP’s customer service.

ISPs should be wary of the problems. Analysis of the data suggests that those who are less satisfied with their internet service, measured either by satisfaction with customer service or by willingness to recommend the ISP to their neighbors, are also more likely to be considering switching their service provider (Correlation coefficients of r=.47 and r=0.42, respectively). This also indicates that those who experience poor customer service are slightly more likely to leave their service provider compared with those who would not recommend their internet service provider to a neighbor.

The rise of mobile internet provided by cell phone companies in the Caucasus adds yet another angle to the story and is likely to further drive competition in the internet market, which in theory should spur on service improvement. While mobile internet is not a fully substitutable good for land-line services for everyone, it will soon be for many. On the 2013 Caucasus Barometer (CB) survey, the activity Georgians most frequently report doing online was social network usage, something which can easily be done on even a very modest model of a smart phone. While CB 2013 indicated that only 12% of Georgian households had internet access on their mobile phone, the share has undoubtedly increased since, and will continue to increase, in line with the global trend in smart phone ownership in recent years and taking into consideration relatively low prices of mobile internet in Georgia.

It is important to remember the above presented provisional impressions from a quick poll that does not claim to be representative of internet users in Tbilisi. It would take a full-scale survey (at the price of a new car) to get representative and comprehensive findings. For now, the key take away message appears to be that customers who experience poor service – both in terms of quality of internet provision and customer service – are more likely to think about switching their provider. While service providers would be wise to improve their customer service, competition stemming from the rise of mobile internet will also likely lead to increasing quality of service as providers compete for subscribers. Improving customer service in Georgian language should be a priority for companies like Silknet and Caucasus Online, as Georgian speaking customers generally rate their service worse than those who receive service in English.

What has your experience with the internet in Tbilisi (and beyond, throughout the Caucasus) been? Join in the conversation on the CRRC-Georgia Facebook page.

The opinions expressed in this blog post reflect those of the authors alone, and do not necessarily represent the views of the organizations which Dustin and Hans work for.