Internap's business model has evolved a few times throughout the years.
The Company was born as an IP service provider, buying capacity from several other networks and reselling it as a better - and more expensive - service thanks to its patented route optimization technology.
Later on, after the acquisition of Sockeye Networks and NetVmg, Internap has also started selling route optimization equipment.
Due to excess capacity and decreasing IP prices in the market, this business model has never been able to achieve any profitability.
The colocation part of the business, partially offered through owned data centers and partially through sites run by specialised partners like Equinix, Switch anda Data, etc. , has been the main growth driver in the last few months.
Although colocation has been offered in the past almost at cost to support the sale of IP services, Internap has been able to grow margins in this sector, as well, and now customers filling its own data centers should guarantee gross profit margins in the 40% range.
CDN (and ad insertion technology) is the latest addition to the product mix, through the acquisition of Vitalstream.
Internap has always enjoyed a very good reputation among its customers for the quality of its services, reaching some of the best names in the financial, technology, government and media sectors.
The CEO change has been a great positive for the Company.
The market cap is probably lower than the sum of all its parts (IP Services, colocation, CDN).
The addition of CDN services and the strength in colocation has finally allowed the Company to grow revenues at an accelerated pace, while keeping good margins.
Internap has no customer representing more than 2% of its revenues, and this can be very positive in some area of business like CDN or colocation, where customers might end up opting to build CDN internally or to look for large data center footprints with wholesalers or REITS.
Margins - Internap enjoys great margins and should be able to maintain this position due to a combination of factors - The Company is buying bandwidth in large scale, so negotiating very good pricing, growing the colocation business having finally reached the critical mass necessary to cover all fixed costs (owned data centers, mainly), and has few big customers that might have access to large disconts.
Internap is not and probably will never be a real leader in any of its fields (IP Services, colocation, CDN). Its mixed business model is probably not enough attractive for the largest customers, who will probably chose Equinix for colocation, Level 3 as their main bandwidth provider, or Akamai for large CDN deals, just as an example. This might also justify the low ARPU (compared to Equinix or Akamai).
Analysts will find it difficult to evaluate Internap because of its mixed business model and multiples will never have the premium enjoyed by the leaders in each field.
There have been many management changes in the team, last one the loss of the head of sales.
Internap hasn't got a great record when it comes to integrating with other Companies.
Internap business model is built around "intelligent routing". Apart from a (relatively) small group of customers (say financial institutions, etc.) who are VERY sensible to performance, this is not a main issue today. If there will be a capacity issue , Internap might find itself in the right spot to finally see its sales accelerate growth.