I love graphs which go upwards. What did I learn from this one? It's the graph of traffic to my own personal website from just one mobile device – the Apple iPhone – over the past year. There are similar graphs for traffic from iPods as well as Android, BlackBerry and SymbianOS devices.

Windows, Mac and Linux computers represented 98% of traffic in August 2009, but only 90% in August 2010. OK, that means the vast majority of visitors are still using PCs with (presumably) big screens, but that growing proportion which has just hit 10% is very significant. It's over 1,700 visitors a month, and if my personal website was a commercial operation, that could mean serious business opportunities.

Fortunately, most of my site looks great on small screens (it has a separate stylesheet and looks totally different). But does yours? Have you even looked at it on a mobile phone? It's increasingly important that it looks good on more than just a desktop Windows PC.

Sometimes you can't see the wood for the trees, and in working hard running Google AdWords campaigns for so many industrial and scientific companies, I will admit to not really having appreciated the inexorable rise in the cost of this essential advertising medium. Of course, it was always this way: in the past when we all advertised in print magazines, the more successful a magazine became, the thicker it got and the less likely our ad was to be seen. Less for the same money. Over the past five years, this effect has been happening with Google AdWords, and today it's rare that you'll get a decent position on a traffic-generating search term for a bid of under £1 a click. As recently as a couple of years ago, I'd see 25 pence clicks quite frequently. But no longer.

That said, Google AdWords is still, without question, the best place to be spending your advertising budget, because as the cost of this medium has slowly risen, others (online and off) have fallen in effectiveness …but not cost. Even if you completely outsource your campaigns to organisations like ours, and end up paying a fixed rate of – typically – £1.50 a click, you'll still get, for example, 500 visitors for a £750 advertising spend, and if you can convert them to enquiries at 10%, that's 50 enquiries. Show me anywhere else you can get sales leads for industrial and scientific products at £15 a time, reliably, and I'll be very interested.

So Google AdWords: a great value medium, but don't think it's a cheap one.

Most companies like to structure their websites in a hierarchy, with separate pages on each individual product at the bottom of the pyramid. If you're starting again with your website, there are many arguments against doing it this way, but I'm going to assume you're stuck with that way of doing things for now, and would like to make the best of it.

What gets pages found in Google is content. Plenty of words. That's why there's a problem with having a separate page for each product, because many of the pages are going to be quite similar, and you don't want to repeat yourself. If you can amalgamate some of these pages, so much the better. But regardless, take a look at your product pages, and ask yourself: "is there enough content on this page to make Google consider it to be an authoritative page for blue widgets?"

The chances are that there's not much content – a brief description and maybe a few specs wouldn't be unusual. What you really need is some good meaty background about the technology. Sure, many visitors will know how a blue widget works, and why it's better than the old-fashioned red type. But they won't think any less of you for explaining that again. (This is quite topical here at BMON, because two clients have recently commissioned us to go right through their website and produce additional background copy for each product range.)

Secondly, how compelling is the headline on the page? Does it say "Blue Widget"? Or does it say "Blue Widget: faster and more reliable performance than a Red Widget"? Is it selling the product? If not, why not?

Thirdly, how appealing is the product shot? Is there just one, scanned off the datasheet? Would you really put that in a brochure for the product? This page will be the first (and potentially only) time many customers will come across the product, and it needs to make an impact. It doesn't cost any more to use multiple images, showing the product inside and out.

Finally, is there a direct call-to-action, such as a form which visitors can complete to get a brochure in the post, or any other response? Telling them to visit the site's all-purpose "contact us" form and fill in what product they want information on isn't very clever. Having no call-to-action at all is even worse.

Plenty of informative content, a great headline, good product images and a call-to-action will get people to your product pages and then acting on what they see. What are your pages like?

A client recently – rather hesitantly – queried something I'd suggested to them, as if it was more likely they'd misunderstood it than I'd got it "wrong". They'd read an article which seemed to contradict what I'd said, and found that surprising. It shouldn't have been. In online marketing we're dealing with something that has no rules, never mind regulations. When it comes to search engine marketing, that's even more the case. There are things which appear to work well in Google, and others which appear to cause problems. But if I say "this works well", there'll always be examples of people not doing it and still getting good results. If I say "avoid this", there'll always be those who go ahead and do it, without any apparent penalty. People are less confident with online marketing than they have been "offline", and would like a Highway Code. Unfortunately there are only unofficial guidebooks.

One thing to remember is that Google is just a commercial entity. Many people treat it as if, online, it's the law. I once listened to someone explain how he made a lot of money online by creating a million websites overnight to swamp Google's results; rather than admire his ingenuity, people in the room were horrified, as if it was illegal. Do what you need to do to make your online marketing successful, and don't be scared of trying anything …as long as it's legal, decent, inoffensive and truthful.

As a magazine and website editor for nearly twenty years, I had to read more industrial and scientific product marketing material than it's healthy for one individual to suffer. In the mid nineties, it took me two hours a day to go through the post, which was 95% press releases. Ten years later, I was receiving even more stuff, but at least the 100 or more submissions a day were coming in via email then. In all that time, the one thing which never failed to irritate me was the indiscriminate use of the word "solution". The word "product" seemed to drop out of favour in the mid-eighties, probably about the same time it was appropriated by the financial services "industry", which started to sell financial "products", as if its services were something tangible. That term was eventually taken over by the beauty "industry", and nowadays "products" (or worse, "product") refers to something you put in your hair.

But back to "solutions", which of course you can also put in your hair if you add water to a concentrate. I don't have any problem with anyone having a solution to my problem. Where I get uncomfortable is when people are offering solutions to non-existent (or at least unperceived) problems. Seth Godin blogged about this recently. Just because your product can make people's lives easier or cheaper doesn't mean they have a "problem", and if they don't have a problem then you're wasting your time promoting your product as a solution. The only way this can work is if you first devote some effort to convincing them they do have a problem, and that's not easy.

Web search is a nice way of getting clarity on this issue. Several clients have said to me that they're finding it difficult to come up with search terms which are relevant to their products or services. This is because people aren't looking for what they sell. Their potential customers don't have a problem and aren't searching for a solution. If that's your situation, I'd suggest that search engine marketing isn't for you. There's no point in spending money on Google AdWords or trying to get to the top of the Google search results if people aren't going to be searching for what you sell. A great website is needed, but you're going to have to alert people to the advantages you can offer them by other methods.

This is quite lovely, and a bit of fun as it's a bank holiday today. Icons of the Web is a representation of the most-visited sites on the web by their icons (the little ones which appear next to the URL in your browser's address bar). The creators scanned the top 1,000,000 sites and collected nearly 300,000 different icons. Then they added up the traffic to the sites which used each icon, and created this chart with the icons blown up to the scale of their websites' traffic (which I like to think of as the number of times they're seen).

Sadly this blog hasn't made it yet, but there are one or two BMON clients I see in there, and I'm sure if you feature you'll want to request the poster for your office. An interactive element on the NMAP site allows you to see if your site's there and zoom in on it. I'd love to see a version from 2000, to compare it with today (Google probably wouldn't have featured back then, for example).

As an aside, your site does have its own icon (AKA "favicon"), doesn't it?

In yesterday's post about Google AdWords, I hinted that one of the main problems advertisers have is not noticing that their ads are running against completely off-target searches. There are many ways to keep an eye on this, but the main ones are featured in a good article on Ask Enquiro called Viewing AdWords Search Query Terms: Let Us Count the Ways!. In particular, the "Search Query Performance report" is now very good, although if you've been advertising with AdWords for a while and not really kept on top of when your ad has been showing, you may want to shield your eyes.

One of the things about Google AdWords advertising is that it takes a lot of your time, which is why so many people subcontract it. However, many of the companies you get phone calls from offering to do this are, quite frankly, rubbish. Ask them just one question: what do you know about my technology?

The answer will almost invariably be "nothing", although they'll weasel around the question and try to pretend this doesn't matter. But it's critical. Anyone setting up an AdWords campaign has to develop a long list of search terms to advertise against, and if they don't know their blue widgets from their red widgets, your campaign will perform badly and cost you much more than it should.

In the past year, we've taken over the running of several companies' AdWords campaigns from conventional advertising agencies. In almost every case, what we've found ranges from a campaign whose creator clearly can't be bothered, to those which are breathtakingly incompetent. This is not because the agency doesn't understand Google AdWords advertising, but because they don't understand the client's area of technology. We get the job – eventually – because we know industrial and scientific marketing.

For example, supposing you sold carbon dioxide canisters. What search terms would you advertise against? The Google AdWords bucket shop would advertise against the search term "carbon dioxide canisters". You and I would advertise against "carbon dioxide canisters" …and "co2 canisters". Pretty obvious to anyone with any technical sense, but not to the sort of people whose main skill is in phoning round lots of businesses and selling advertising. Or suppose you sold industrial AC and DC motors? (I've seen this one, by the way). An agency who didn't know the subject might just advertise against "motors". The result? Your advert shows millions of times against queries like "General Motors dealers". And you get the bill for any clickthroughs.

Google AdWords isn't really that difficult to master, if you know your subject …and nobody knows it better than you. The sensible reason for subcontracting it out is because you don't have the time to manage it properly. If that's the case, then give the task to someone who talks your language, not a call-centre full of people who neither know or care about what you do.

If you're interested in Google AdWords advertising (especially if you might consider asking us to do it!), we have a free seminar coming upon 21 October. More details here.

If you've not used Google Docs, I'd recommend taking a look. It's basically a set of "office" applications (word processor, spreadsheet etc) but with the software all online, so you don't need to have any programs on your computer. For the first time ever, I didn't buy Microsoft Office with a recent PC purchase, because I can do everything using Google Docs, thanks. It does require a permanent internet connection, but that's not a problem, I wasn't planning to go on the road with this PC. Also, the documents are stored at Google (your PC really is just a window onto another computer), which I vastly prefer to having them stored locally.

There are no more software upgrades because Google has the application and therefore takes care of that, and – slowly – it's not only catching up with conventional Microsoft Office, but is surpassing it. A recent introduction, for example, was automatic document translation, which is wonderfully convenient. Machine translation is far from perfect, but I often find if I've asked a client or colleague to translate something, and they get a machine translation to accompany the English one, it rather focuses their mind on the task, for fear that if they don't do it, the machine translation will be used. And for translating foreign language documents into English, if you're doing research, it's an awesome tool.

There are many ways of researching the important search terms for your area of business, notably the Google AdWords Keyword Tool, but one of the easiest is just to type something into Google and look at the suggestions it produces. If you're just brainstorming ideas, it's a useful way of kicking things off.

Obviously this is not scientific, just a way of getting the ball rolling, and sometimes there are some quite bizarre suggestions. I will, of course, appreciate receiving screenshots of any of those which you see.