Reducing Long-Distance PSTN Costs at Cisco
In early 2009, I was asked to form a virtual, global team within Cisco IT to figure out how we could reduce international long distance calling expenses at Cisco. As a result of adding 28 TEHO routes we were able to avoid future PSTN costs of approximately $4 million per year.
Years ago we had already started setting up a call-routing technique known as “tail-end hop off” or TEHO, and over time our IP Telephony team had implemented TEHO to 31 countries on some, but not all, clusters worldwide. A lot of people didn’t think that there was any more savings we could get out of our TEHO plan, but we decided to review it again anyway, and it’s a good thing we did. In just a few years our calling patterns had changed a lot – thanks to our expansion in different parts of Europe, the Middle East, Asia and India. Also, our long distance tariffs from our larger sites where we typically transport high volumes of telecom traffic had changed a lot since we had last looked. So our team’s challenge was to identify better routing plans to take advantage of tariff changes, and more sites where TEHO would be cost-effective.
In the end we realized that there are five major issues to consider when choosing new countries for us to expand our TEHO call routing.
- The country would have a high volume of international calls terminating on external numbers in that country, and our bills from other offices around the globe would show high calling costs to that particular country.
- Local regulations would not prohibit or significantly hinder our ability to implement TEHO (there are a lot of countries in the Middle East and Asia where TEHO is not permitted).
- For sites where we could see an opportunity for savings, care need to be taken to ensure that WAN and PSTN resources for the site would not be oversubscribed by the additional traffic. This was especially relevant for PSTN resources as we needed to ensure both in and outbound connectivity for the office itself. If additional resources were required, the planned savings had to accommodate the additional infrastructure costs as well as a viable savings to warrant the effort. (For TEHO calls that would terminate out of local trunks, not all sites had enough trunks to accommodate the expected increase in the call load that TEHO would bring.)
- We had to add in the costs and issues that might arise for operational support for these trunks, and for monitoring voice quality at the site.
- Lastly, for countries where we had identified significant TEHO opportunities but the local office at that location had insufficient resources we looked towards a comparison between international tariffs from our main providers at our largest sites. This analysis highlighted some major differences in call costs. By choosing to aggregate our traffic from which ever was the cheaper carrier to those limited resources countries identified for TEHO we were able to overcome the resource issue (with no extra investment) and still make considerable savings.
In addition to the issues identified above we created a simple check list to cover off what was required in order to facilitate TEHO at either a local hop-off or aggregated hop-off point. These steps can be applied to any scenario in which TEHO is to be considered and cover both initial analysis, tracking of savings and network impact:
step 1 – If the change is TEHO (Tail End Hop-Off) related, check for legal clarification before moving forward.
step 2 – Assess the location and impact on WAN Bandwidth for the suggested change? (incl. increase and/or decrease)
step 3 – Assess the location and impact on PSTN Circuits for the suggested change? (incl. increase and/or decrease)
step 4 – Establish baseline from which to measure BW and Operating Expense impacts? This may involve recording the PSTN costs prior to the change and Bandwidth usage (both current and planned).
step 5 – Establish departments and/or business organizations that will benefit from decreased cost impacts of Routing changes?
step 6 – Establish departments and/or business organizations that will see increases in cost due to Routing changes?
step 7 – Establish the increase or decrease in terms of operational support?
step 8 – Conduct voice tests for the proposed configuration. These must include voice quality and latency tests.
step 10 – Notifications of the change must be sent to all relevant Network teams for awareness prior to changes being implemented.
step 11 – Implement voice routing changes. Dates of changes in all regions must be supplied to those responsible for tracking savings and capacity management.
Based upon the evaluation that was carried out as part of this virtual teams charter, we chose 28 new countries for TEHO, which we expect will allow us to avoid future costs of at least $4 million annually.
Here is another tip for reducing voice-call costs. Be sure to compare the costs of an international long-distance call to a local call at each site. Although it may seem illogical, local per-minute tariffs may be more expensive than long-distance rates that reflect a discount based on high corporate call volumes.
- Case Study: Cisco IT Voice Call Routing Case Study
Troy Wylie-Hill is a Cisco IT network engineer in Amsterdam